diff --git a/Cargo.lock b/Cargo.lock
index 56ba0f14c..d6628322b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -130,7 +130,7 @@ dependencies = [
  "aws-http",
  "aws-runtime",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
@@ -181,9 +181,9 @@ dependencies = [
 
 [[package]]
 name = "anstream"
-version = "0.6.19"
+version = "0.6.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "301af1932e46185686725e0fad2f8f2aa7da69dd70bf6ecc44d6b703844a3933"
+checksum = "3ae563653d1938f79b1ab1b5e668c87c76a9930414574a6583a7b7e11a8e6192"
 dependencies = [
  "anstyle",
  "anstyle-parse",
@@ -211,22 +211,22 @@ dependencies = [
 
 [[package]]
 name = "anstyle-query"
-version = "1.1.3"
+version = "1.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6c8bdeb6047d8983be085bab0ba1472e6dc604e7041dbf6fcd5e71523014fae9"
+checksum = "9e231f6134f61b71076a3eab506c379d4f36122f2af15a9ff04415ea4c3339e2"
 dependencies = [
- "windows-sys 0.59.0",
+ "windows-sys 0.60.2",
 ]
 
 [[package]]
 name = "anstyle-wincon"
-version = "3.0.9"
+version = "3.0.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "403f75924867bb1033c59fbf0797484329750cfbe3c4325cd33127941fabc882"
+checksum = "3e0633414522a32ffaac8ac6cc8f748e090c5717661fddeea04219e2344f5f2a"
 dependencies = [
  "anstyle",
  "once_cell_polyfill",
- "windows-sys 0.59.0",
+ "windows-sys 0.60.2",
 ]
 
 [[package]]
@@ -246,9 +246,9 @@ dependencies = [
 
 [[package]]
 name = "arboard"
-version = "3.5.0"
+version = "3.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c1df21f715862ede32a0c525ce2ca4d52626bb0007f8c18b87a384503ac33e70"
+checksum = "55f533f8e0af236ffe5eb979b99381df3258853f00ba2e44b6e1955292c75227"
 dependencies = [
  "clipboard-win",
  "log",
@@ -295,9 +295,9 @@ dependencies = [
 
 [[package]]
 name = "async-compression"
-version = "0.4.25"
+version = "0.4.27"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "40f6024f3f856663b45fd0c9b6f2024034a702f453549449e0d84a305900dad4"
+checksum = "ddb939d66e4ae03cee6091612804ba446b12878410cfa17f785f4dd67d4014e8"
 dependencies = [
  "flate2",
  "futures-core",
@@ -331,9 +331,9 @@ checksum = "c08606f8c3cbf4ce6ec8e28fb0014a2c086708fe954eaa885384a6165172e7e8"
 
 [[package]]
 name = "aws-config"
-version = "1.8.0"
+version = "1.8.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "455e9fb7743c6f6267eb2830ccc08686fbb3d13c9a689369562fd4d4ef9ea462"
+checksum = "483020b893cdef3d89637e428d588650c71cfae7ea2e6ecbaee4de4ff99fb2dd"
 dependencies = [
  "aws-credential-types",
  "aws-runtime",
@@ -341,7 +341,7 @@ dependencies = [
  "aws-sdk-ssooidc",
  "aws-sdk-sts",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
@@ -361,9 +361,9 @@ dependencies = [
 
 [[package]]
 name = "aws-credential-types"
-version = "1.2.3"
+version = "1.2.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "687bc16bc431a8533fe0097c7f0182874767f920989d7260950172ae8e3c4465"
+checksum = "1541072f81945fa1251f8795ef6c92c4282d74d59f88498ae7d4bf00f0ebdad9"
 dependencies = [
  "aws-smithy-async",
  "aws-smithy-runtime-api",
@@ -382,9 +382,9 @@ dependencies = [
 
 [[package]]
 name = "aws-lc-rs"
-version = "1.13.1"
+version = "1.13.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "93fcc8f365936c834db5514fc45aee5b1202d677e6b40e48468aaaa8183ca8c7"
+checksum = "5c953fe1ba023e6b7730c0d4b031d06f267f23a46167dcbd40316644b10a17ba"
 dependencies = [
  "aws-lc-sys",
  "zeroize",
@@ -392,9 +392,9 @@ dependencies = [
 
 [[package]]
 name = "aws-lc-sys"
-version = "0.29.0"
+version = "0.30.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "61b1d86e7705efe1be1b569bab41d4fa1e14e220b60a160f78de2db687add079"
+checksum = "dbfd150b5dbdb988bcc8fb1fe787eb6b7ee6180ca24da683b61ea5405f3d43ff"
 dependencies = [
  "bindgen 0.69.5",
  "cc",
@@ -405,14 +405,14 @@ dependencies = [
 
 [[package]]
 name = "aws-runtime"
-version = "1.5.8"
+version = "1.5.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4f6c68419d8ba16d9a7463671593c54f81ba58cab466e9b759418da606dcc2e2"
+checksum = "c034a1bc1d70e16e7f4e4caf7e9f7693e4c9c24cd91cf17c2a0b21abaebc7c8b"
 dependencies = [
  "aws-credential-types",
  "aws-sigv4",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
  "aws-smithy-types",
@@ -429,14 +429,14 @@ dependencies = [
 
 [[package]]
 name = "aws-sdk-cognitoidentity"
-version = "1.74.0"
+version = "1.80.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2a14e1e7fbff8dccd5ad97c52478c888bd8ccc70206d06b027db2eb4904c4a24"
+checksum = "606ea1307a8caf6ecddf1d8bbe55b7d1d967c7b6c217d13d5e81c2c8e24e0cf8"
 dependencies = [
  "aws-credential-types",
  "aws-runtime",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
@@ -451,14 +451,14 @@ dependencies = [
 
 [[package]]
 name = "aws-sdk-sso"
-version = "1.73.0"
+version = "1.79.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b2ac1674cba7872061a29baaf02209fefe499ff034dfd91bd4cc59e4d7741489"
+checksum = "0a847168f15b46329fa32c7aca4e4f1a2e072f9b422f0adb19756f2e1457f111"
 dependencies = [
  "aws-credential-types",
  "aws-runtime",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
@@ -473,14 +473,14 @@ dependencies = [
 
 [[package]]
 name = "aws-sdk-ssooidc"
-version = "1.74.0"
+version = "1.80.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3a6a22f077f5fd3e3c0270d4e1a110346cddf6769e9433eb9e6daceb4ca3b149"
+checksum = "b654dd24d65568738593e8239aef279a86a15374ec926ae8714e2d7245f34149"
 dependencies = [
  "aws-credential-types",
  "aws-runtime",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-runtime",
  "aws-smithy-runtime-api",
@@ -495,14 +495,14 @@ dependencies = [
 
 [[package]]
 name = "aws-sdk-sts"
-version = "1.75.0"
+version = "1.81.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e3258fa707f2f585ee3049d9550954b959002abd59176975150a01d5cf38ae3f"
+checksum = "c92ea8a7602321c83615c82b408820ad54280fb026e92de0eeea937342fafa24"
 dependencies = [
  "aws-credential-types",
  "aws-runtime",
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-json",
  "aws-smithy-query",
  "aws-smithy-runtime",
@@ -518,12 +518,12 @@ dependencies = [
 
 [[package]]
 name = "aws-sigv4"
-version = "1.3.3"
+version = "1.3.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ddfb9021f581b71870a17eac25b52335b82211cdc092e02b6876b2bcefa61666"
+checksum = "084c34162187d39e3740cb635acd73c4e3a551a36146ad6fe8883c929c9f876c"
 dependencies = [
  "aws-credential-types",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-runtime-api",
  "aws-smithy-types",
  "bytes",
@@ -551,9 +551,9 @@ dependencies = [
 
 [[package]]
 name = "aws-smithy-eventstream"
-version = "0.60.9"
+version = "0.60.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "338a3642c399c0a5d157648426110e199ca7fd1c689cc395676b81aa563700c4"
+checksum = "604c7aec361252b8f1c871a7641d5e0ba3a7f5a586e51b66bc9510a5519594d9"
 dependencies = [
  "aws-smithy-types",
  "bytes",
@@ -583,9 +583,9 @@ dependencies = [
 
 [[package]]
 name = "aws-smithy-http"
-version = "0.62.1"
+version = "0.62.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "99335bec6cdc50a346fda1437f9fefe33abf8c99060739a546a16457f2862ca9"
+checksum = "7c4dacf2d38996cf729f55e7a762b30918229917eca115de45dfa8dfb97796c9"
 dependencies = [
  "aws-smithy-runtime-api",
  "aws-smithy-types",
@@ -603,17 +603,17 @@ dependencies = [
 
 [[package]]
 name = "aws-smithy-http-client"
-version = "1.0.5"
+version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7f491388e741b7ca73b24130ff464c1478acc34d5b331b7dd0a2ee4643595a15"
+checksum = "f108f1ca850f3feef3009bdcc977be201bca9a91058864d9de0684e64514bee0"
 dependencies = [
  "aws-smithy-async",
  "aws-smithy-protocol-test",
  "aws-smithy-runtime-api",
  "aws-smithy-types",
  "bytes",
- "h2 0.3.26",
- "h2 0.4.10",
+ "h2 0.3.27",
+ "h2 0.4.12",
  "http 0.2.12",
  "http 1.3.1",
  "http-body 0.4.6",
@@ -626,7 +626,7 @@ dependencies = [
  "indexmap",
  "pin-project-lite",
  "rustls 0.21.12",
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "rustls-native-certs 0.8.1",
  "rustls-pki-types",
  "serde",
@@ -685,12 +685,12 @@ dependencies = [
 
 [[package]]
 name = "aws-smithy-runtime"
-version = "1.8.3"
+version = "1.8.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "14302f06d1d5b7d333fd819943075b13d27c7700b414f574c3c35859bfb55d5e"
+checksum = "9e107ce0783019dbff59b3a244aa0c114e4a8c9d93498af9162608cd5474e796"
 dependencies = [
  "aws-smithy-async",
- "aws-smithy-http 0.62.1",
+ "aws-smithy-http 0.62.3",
  "aws-smithy-http-client",
  "aws-smithy-observability",
  "aws-smithy-runtime-api",
@@ -710,9 +710,9 @@ dependencies = [
 
 [[package]]
 name = "aws-smithy-runtime-api"
-version = "1.8.1"
+version = "1.8.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bd8531b6d8882fd8f48f82a9754e682e29dd44cff27154af51fa3eb730f59efb"
+checksum = "75d52251ed4b9776a3e8487b2a01ac915f73b2da3af8fc1e77e0fce697a550d4"
 dependencies = [
  "aws-smithy-async",
  "aws-smithy-types",
@@ -775,9 +775,9 @@ dependencies = [
 
 [[package]]
 name = "aws-types"
-version = "1.3.7"
+version = "1.3.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8a322fec39e4df22777ed3ad8ea868ac2f94cd15e1a55f6ee8d8d6305057689a"
+checksum = "b069d19bf01e46298eaedd7c6f283fe565a59263e53eebec945f3e6398f42390"
 dependencies = [
  "aws-credential-types",
  "aws-smithy-async",
@@ -948,9 +948,9 @@ dependencies = [
 
 [[package]]
 name = "bm25"
-version = "2.2.1"
+version = "2.3.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9874599901ae2aaa19b1485145be2fa4e9af42d1b127672a03a7099ab6350bac"
+checksum = "b84ff0d57042bc263e2ebadb3703424b59b65870902649a2b3d0f4d7ab863244"
 dependencies = [
  "cached",
  "deunicode",
@@ -989,9 +989,9 @@ dependencies = [
 
 [[package]]
 name = "bumpalo"
-version = "3.18.1"
+version = "3.19.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "793db76d6187cd04dff33004d8e6c9cc4e05cd330500379d2394209271b4aeee"
+checksum = "46c5e41b57b8bba42a04676d81cb89e9ee8e859a1a66f80a5a72e1cb76b34d43"
 
 [[package]]
 name = "bytecount"
@@ -1001,18 +1001,18 @@ checksum = "175812e0be2bccb6abe50bb8d566126198344f707e304f45c648fd8f2cc0365e"
 
 [[package]]
 name = "bytemuck"
-version = "1.23.1"
+version = "1.23.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5c76a5792e44e4abe34d3abf15636779261d45a7450612059293d1d2cfc63422"
+checksum = "3995eaeebcdf32f91f980d360f78732ddc061097ab4e39991ae7a6ace9194677"
 dependencies = [
  "bytemuck_derive",
 ]
 
 [[package]]
 name = "bytemuck_derive"
-version = "1.9.3"
+version = "1.10.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7ecc273b49b3205b83d648f0690daa588925572cc5063745bfe547fe7ec8e1a1"
+checksum = "4f154e572231cb6ba2bd1176980827e3d5dc04cc183a75dea38109fbdd672d29"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -1043,14 +1043,14 @@ dependencies = [
 
 [[package]]
 name = "cached"
-version = "0.55.1"
+version = "0.56.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b0839c297f8783316fcca9d90344424e968395413f0662a5481f79c6648bbc14"
+checksum = "801927ee168e17809ab8901d9f01f700cd7d8d6a6527997fee44e4b0327a253c"
 dependencies = [
  "ahash",
  "cached_proc_macro",
  "cached_proc_macro_types",
- "hashbrown 0.14.5",
+ "hashbrown 0.15.5",
  "once_cell",
  "thiserror 2.0.12",
  "web-time",
@@ -1058,9 +1058,9 @@ dependencies = [
 
 [[package]]
 name = "cached_proc_macro"
-version = "0.24.0"
+version = "0.25.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "673992d934f0711b68ebb3e1b79cdc4be31634b37c98f26867ced0438ca5c603"
+checksum = "9225bdcf4e4a9a4c08bf16607908eb2fbf746828d5e0b5e019726dbf6571f201"
 dependencies = [
  "darling",
  "proc-macro2",
@@ -1076,9 +1076,9 @@ checksum = "ade8366b8bd5ba243f0a58f036cc0ca8a2f069cff1a2351ef1cac6b083e16fc0"
 
 [[package]]
 name = "camino"
-version = "1.1.10"
+version = "1.1.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0da45bc31171d8d6960122e222a67740df867c1dd53b4d51caa297084c185cab"
+checksum = "5d07aa9a93b00c76f71bc35d598bed923f6d4f3a9ca5c24b7737ae1a292841c0"
 dependencies = [
  "serde",
 ]
@@ -1095,7 +1095,7 @@ dependencies = [
  "memmap2",
  "num-traits",
  "num_cpus",
- "rand 0.9.1",
+ "rand 0.9.2",
  "rand_distr",
  "rayon",
  "safetensors",
@@ -1131,7 +1131,7 @@ dependencies = [
  "candle-nn",
  "fancy-regex 0.13.0",
  "num-traits",
- "rand 0.9.1",
+ "rand 0.9.2",
  "rayon",
  "serde",
  "serde_json",
@@ -1145,6 +1145,15 @@ version = "0.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "37b2a672a2cb129a2e41c10b1224bb368f9f37a2b16b612598138befd7b37eb5"
 
+[[package]]
+name = "castaway"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dec551ab6e7578819132c713a93c022a05d60159dc86e7a7050223577484c55a"
+dependencies = [
+ "rustversion",
+]
+
 [[package]]
 name = "cbor-diag"
 version = "0.1.12"
@@ -1166,9 +1175,9 @@ dependencies = [
 
 [[package]]
 name = "cc"
-version = "1.2.27"
+version = "1.2.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d487aa071b5f64da6f19a3e848e3578944b726ee5a4854b82172f02aa876bfdc"
+checksum = "2352e5597e9c544d5e6d9c95190d5d27738ade584fa8db0a16e130e5c2b5296e"
 dependencies = [
  "jobserver",
  "libc",
@@ -1267,17 +1276,17 @@ dependencies = [
  "quote",
  "r2d2",
  "r2d2_sqlite",
- "rand 0.9.1",
+ "rand 0.9.2",
  "regex",
  "reqwest",
  "ring",
  "rusqlite",
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "rustls-native-certs 0.8.1",
  "rustls-pemfile 2.2.0",
  "rustyline",
  "schemars",
- "security-framework 3.2.0",
+ "security-framework 3.3.0",
  "semantic_search_client",
  "semver",
  "serde",
@@ -1291,7 +1300,7 @@ dependencies = [
  "skim",
  "spinners",
  "strip-ansi-escapes",
- "strum 0.27.1",
+ "strum 0.27.2",
  "syn 2.0.104",
  "syntect",
  "sysinfo",
@@ -1383,9 +1392,9 @@ dependencies = [
 
 [[package]]
 name = "clap"
-version = "4.5.40"
+version = "4.5.43"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "40b6887a1d8685cebccf115538db5c0efe625ccac9696ad45c409d96566e910f"
+checksum = "50fd97c9dc2399518aa331917ac6f274280ec5eb34e555dd291899745c48ec6f"
 dependencies = [
  "clap_builder",
  "clap_derive",
@@ -1393,9 +1402,9 @@ dependencies = [
 
 [[package]]
 name = "clap_builder"
-version = "4.5.40"
+version = "4.5.43"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e0c66c08ce9f0c698cbce5c0279d0bb6ac936d8674174fe48f736533b964f59e"
+checksum = "c35b5830294e1fa0462034af85cc95225a4cb07092c088c55bda3147cfcd8f65"
 dependencies = [
  "anstream",
  "anstyle",
@@ -1408,9 +1417,9 @@ dependencies = [
 
 [[package]]
 name = "clap_complete"
-version = "4.5.54"
+version = "4.5.56"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "aad5b1b4de04fead402672b48897030eec1f3bfe1550776322f59f6d6e6a5677"
+checksum = "67e4efcbb5da11a92e8a609233aa1e8a7d91e38de0be865f016d14700d45a7fd"
 dependencies = [
  "clap",
 ]
@@ -1427,9 +1436,9 @@ dependencies = [
 
 [[package]]
 name = "clap_derive"
-version = "4.5.40"
+version = "4.5.41"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d2c7947ae4cc3d851207c1adb5b5e260ff0cca11446b1d6d1423788e442257ce"
+checksum = "ef4f52386a59ca4c860f7393bcf8abd8dfd91ecccc0f774635ff68e92eeef491"
 dependencies = [
  "anstyle",
  "heck 0.5.0",
@@ -1447,9 +1456,9 @@ checksum = "b94f61472cee1439c0b966b47e3aca9ae07e45d070759512cd390ea2bebc6675"
 
 [[package]]
 name = "clipboard-win"
-version = "5.4.0"
+version = "5.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "15efe7a882b08f34e38556b14f2fb3daa98769d06c7f0c1b076dfd0d983bc892"
+checksum = "bde03770d3df201d4fb868f2c9c59e66a3e4e2bd06692a0fe701e7103c7e84d4"
 dependencies = [
  "error-code",
 ]
@@ -1536,6 +1545,21 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "compact_str"
+version = "0.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fdb1325a1cece981e8a296ab8f0f9b63ae357bd0784a9faaf548cc7b480707a"
+dependencies = [
+ "castaway",
+ "cfg-if",
+ "itoa",
+ "rustversion",
+ "ryu",
+ "serde",
+ "static_assertions",
+]
+
 [[package]]
 name = "console"
 version = "0.15.11"
@@ -1634,9 +1658,9 @@ dependencies = [
 
 [[package]]
 name = "crc32fast"
-version = "1.4.2"
+version = "1.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a97769d94ddab943e4510d138150169a2758b5ef3eb191a9ee688de3e23ef7b3"
+checksum = "9481c1c90cbf2ac953f07c8d4a58aa3945c425b7185c9154d67a65e4230da511"
 dependencies = [
  "cfg-if",
 ]
@@ -1758,9 +1782,9 @@ dependencies = [
 
 [[package]]
 name = "crunchy"
-version = "0.2.3"
+version = "0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "43da5946c66ffcc7745f48db692ffbb10a83bfe0afd96235c5c2a4fb23994929"
+checksum = "460fbee9c2c2f33933d720630a6a0bac33ba7053db5344fac858d4b8952d77d5"
 
 [[package]]
 name = "crypto-common"
@@ -1817,6 +1841,15 @@ dependencies = [
  "syn 2.0.104",
 ]
 
+[[package]]
+name = "dary_heap"
+version = "0.3.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "04d2cd9c18b9f454ed67da600630b021a8a80bf33f8c95896ab33aaf1c26b728"
+dependencies = [
+ "serde",
+]
+
 [[package]]
 name = "data-encoding"
 version = "2.9.0"
@@ -1976,7 +2009,7 @@ checksum = "e01a3366d27ee9890022452ee61b2b63a67e6f13f58900b651ff5665f0bb1fab"
 dependencies = [
  "libc",
  "option-ext",
- "redox_users 0.5.0",
+ "redox_users 0.5.2",
  "windows-sys 0.60.2",
 ]
 
@@ -2041,9 +2074,9 @@ checksum = "92773504d58c093f6de2459af4af33faa518c13451eb8f2b5698ed3d36e7c813"
 
 [[package]]
 name = "dyn-clone"
-version = "1.0.19"
+version = "1.0.20"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1c7a8fb8a9fbf66c1f703fe16184d10ca0ee9d23be5b4436400408ba54a95005"
+checksum = "d0881ea181b1df73ff77ffaaf9c7544ecc11e82fba9b5f27b262a3c73a332555"
 
 [[package]]
 name = "dyn-stack"
@@ -2252,7 +2285,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0ce92ff622d6dadf7349484f42c93271a0d49b7cc4d466a936405bacbe10aa78"
 dependencies = [
  "cfg-if",
- "rustix 1.0.7",
+ "rustix 1.0.8",
  "windows-sys 0.59.0",
 ]
 
@@ -2748,9 +2781,9 @@ dependencies = [
 
 [[package]]
 name = "h2"
-version = "0.3.26"
+version = "0.3.27"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "81fe527a889e1532da5c525686d96d4c2e74cdd345badf8dfef9f6b39dd5f5e8"
+checksum = "0beca50380b1fc32983fc1cb4587bfa4bb9e78fc259aad4a0032d2080309222d"
 dependencies = [
  "bytes",
  "fnv",
@@ -2767,9 +2800,9 @@ dependencies = [
 
 [[package]]
 name = "h2"
-version = "0.4.10"
+version = "0.4.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a9421a676d1b147b16b82c9225157dc629087ef8ec4d5e2960f9437a90dac0a5"
+checksum = "f3c0b69cfcb4e1b9f1bf2f53f95f766e4661169728ec61cd3fe5a0166f2d1386"
 dependencies = [
  "atomic-waker",
  "bytes",
@@ -2794,7 +2827,7 @@ dependencies = [
  "cfg-if",
  "crunchy",
  "num-traits",
- "rand 0.9.1",
+ "rand 0.9.2",
  "rand_distr",
 ]
 
@@ -2810,9 +2843,9 @@ dependencies = [
 
 [[package]]
 name = "hashbrown"
-version = "0.15.4"
+version = "0.15.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5971ac85611da7067dbfcabef3c70ebb5606018acd9e2a3903a0da507521e0d5"
+checksum = "9229cfe53dfd69f0609a49f65461bd93001ea1ef889cd5529dd176593f5338a1"
 dependencies = [
  "allocator-api2",
  "equivalent",
@@ -2979,14 +3012,14 @@ dependencies = [
  "futures-channel",
  "futures-core",
  "futures-util",
- "h2 0.3.26",
+ "h2 0.3.27",
  "http 0.2.12",
  "http-body 0.4.6",
  "httparse",
  "httpdate",
  "itoa",
  "pin-project-lite",
- "socket2",
+ "socket2 0.5.10",
  "tokio",
  "tower-service",
  "tracing",
@@ -3002,7 +3035,7 @@ dependencies = [
  "bytes",
  "futures-channel",
  "futures-util",
- "h2 0.4.10",
+ "h2 0.4.12",
  "http 1.3.1",
  "http-body 1.0.1",
  "httparse",
@@ -3039,20 +3072,20 @@ dependencies = [
  "http 1.3.1",
  "hyper 1.6.0",
  "hyper-util",
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "rustls-native-certs 0.8.1",
  "rustls-pki-types",
  "tokio",
  "tokio-rustls 0.26.2",
  "tower-service",
- "webpki-roots 1.0.1",
+ "webpki-roots 1.0.2",
 ]
 
 [[package]]
 name = "hyper-util"
-version = "0.1.14"
+version = "0.1.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dc2fdfdbff08affe55bb779f33b053aa1fe5dd5b54c257343c17edfa55711bdb"
+checksum = "8d9b05277c7e8da2c93a568989bb6207bef0112e8d17df7a6eda4a3cf143bc5e"
 dependencies = [
  "base64 0.22.1",
  "bytes",
@@ -3066,7 +3099,7 @@ dependencies = [
  "libc",
  "percent-encoding",
  "pin-project-lite",
- "socket2",
+ "socket2 0.6.0",
  "tokio",
  "tower-service",
  "tracing",
@@ -3211,18 +3244,18 @@ dependencies = [
 
 [[package]]
 name = "indenter"
-version = "0.3.3"
+version = "0.3.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ce23b50ad8242c51a442f3ff322d56b02f08852c77e4c0b4d3fd684abc89c683"
+checksum = "964de6e86d545b246d84badc0fef527924ace5134f30641c203ef52ba83f58d5"
 
 [[package]]
 name = "indexmap"
-version = "2.9.0"
+version = "2.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cea70ddb795996207ad57735b50c5982d8844f38ba9ee5f1aedcfb708a2aa11e"
+checksum = "fe4cd85333e22411419a0bcae1297d25e58c9443848b11dc6a86fefe8c78a661"
 dependencies = [
  "equivalent",
- "hashbrown 0.15.4",
+ "hashbrown 0.15.5",
  "serde",
 ]
 
@@ -3265,6 +3298,17 @@ dependencies = [
  "rustversion",
 ]
 
+[[package]]
+name = "io-uring"
+version = "0.7.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d93587f37623a1a17d94ef2bc9ada592f5465fe7732084ab7beefabe5c77c0c4"
+dependencies = [
+ "bitflags 2.9.1",
+ "cfg-if",
+ "libc",
+]
+
 [[package]]
 name = "ipnet"
 version = "2.11.0"
@@ -3315,27 +3359,27 @@ dependencies = [
 
 [[package]]
 name = "itertools"
-version = "0.11.0"
+version = "0.12.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b1c173a5686ce8bfa551b3563d0c2170bf24ca44da99c7ca4bfdab5418c3fe57"
+checksum = "ba291022dbbd398a455acf126c1e341954079855bc60dfdda641363bd6922569"
 dependencies = [
  "either",
 ]
 
 [[package]]
 name = "itertools"
-version = "0.12.1"
+version = "0.13.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ba291022dbbd398a455acf126c1e341954079855bc60dfdda641363bd6922569"
+checksum = "413ee7dfc52ee1a4949ceeb7dbc8a33f2d6c088194d9f922fb8318faf1f01186"
 dependencies = [
  "either",
 ]
 
 [[package]]
 name = "itertools"
-version = "0.13.0"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "413ee7dfc52ee1a4949ceeb7dbc8a33f2d6c088194d9f922fb8318faf1f01186"
+checksum = "2b192c782037fadd9cfa75548310488aabdbf3d2da73885b31bd0abd03351285"
 dependencies = [
  "either",
 ]
@@ -3442,7 +3486,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "07033963ba89ebaf1584d767badaa2e8fcec21aedea6b8c0346d487d49c28667"
 dependencies = [
  "cfg-if",
- "windows-targets 0.53.2",
+ "windows-targets 0.53.3",
 ]
 
 [[package]]
@@ -3474,12 +3518,13 @@ dependencies = [
 
 [[package]]
 name = "libredox"
-version = "0.1.3"
+version = "0.1.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c0ff37bd590ca25063e35af745c343cb7a0271906fb7b37e4813e8f79f00268d"
+checksum = "391290121bad3d37fbddad76d8f5d1c1c314cfc646d143d7e07a3086ddff0ce3"
 dependencies = [
  "bitflags 2.9.1",
  "libc",
+ "redox_syscall",
 ]
 
 [[package]]
@@ -3531,9 +3576,9 @@ checksum = "241eaef5fd12c88705a01fc1066c48c4b36e0dd4377dcdc7ec3942cea7a69956"
 
 [[package]]
 name = "litrs"
-version = "0.4.1"
+version = "0.4.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b4ce301924b7887e9d637144fdade93f9dfff9b60981d4ac161db09720d39aa5"
+checksum = "f5e54036fe321fd421e10d732f155734c4e4afd610dd556d9a82833ab3ee0bed"
 
 [[package]]
 name = "lock_api"
@@ -3557,7 +3602,7 @@ version = "0.12.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "234cf4f4a04dc1f57e24b96cc0cd600cf2af460d4161ac5ecdd0af8e1f3b2a38"
 dependencies = [
- "hashbrown 0.15.4",
+ "hashbrown 0.15.5",
 ]
 
 [[package]]
@@ -3623,9 +3668,9 @@ checksum = "32a282da65faaf38286cf3be983213fcf1d2e2a58700e808f83f4ea9a4804bc0"
 
 [[package]]
 name = "memmap2"
-version = "0.9.5"
+version = "0.9.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fd3f7eed9d3848f8b98834af67102b720745c4ec028fcd0aa0239277e7de374f"
+checksum = "483758ad303d734cec05e5c12b41d7e93e6a6390c5e9dae6bdeb7c1259012d28"
 dependencies = [
  "libc",
  "stable_deref_trait",
@@ -3743,7 +3788,7 @@ dependencies = [
  "hyper 1.6.0",
  "hyper-util",
  "log",
- "rand 0.9.1",
+ "rand 0.9.2",
  "regex",
  "serde_json",
  "serde_urlencoded",
@@ -4417,9 +4462,9 @@ checksum = "b15813163c1d831bf4a13c3610c05c0d03b39feb07f7e09fa234dac9b15aaf39"
 
 [[package]]
 name = "owo-colors"
-version = "4.2.1"
+version = "4.2.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "26995317201fa17f3656c36716aed4a7c81743a9634ac4c99c0eeda495db0cec"
+checksum = "48dd4f4a2c8405440fd0462561f0e5806bd0f77e86f51c761481bdd4018b545e"
 
 [[package]]
 name = "parking_lot"
@@ -4486,13 +4531,13 @@ checksum = "7edddbd0b52d732b21ad9a5fab5c704c14cd949e5e9a1ec5929a24fded1b904c"
 
 [[package]]
 name = "plist"
-version = "1.7.2"
+version = "1.7.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d77244ce2d584cd84f6a15f86195b8c9b2a0dfbfd817c09e0464244091a58ed"
+checksum = "3af6b589e163c5a788fab00ce0c0366f6efbb9959c2f9874b224936af7fce7e1"
 dependencies = [
  "base64 0.22.1",
  "indexmap",
- "quick-xml",
+ "quick-xml 0.38.1",
  "serde",
  "time",
 ]
@@ -4606,9 +4651,9 @@ dependencies = [
 
 [[package]]
 name = "prettyplease"
-version = "0.2.35"
+version = "0.2.36"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "061c1221631e079b26479d25bbf2275bfe5917ae8419cd7e34f13bfc2aa7539a"
+checksum = "ff24dfcda44452b9816fff4cd4227e1bb73ff5a2f1bc1105aa92fb8565ce44d2"
 dependencies = [
  "proc-macro2",
  "syn 2.0.104",
@@ -4647,9 +4692,9 @@ dependencies = [
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.95"
+version = "1.0.96"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "02b3e5e68a3a1a02aad3ec490a98007cbc13c37cbe84a3cd7b8e406d76e7f778"
+checksum = "beef09f85ae72cea1ef96ba6870c51e6382ebfa4f0e85b643459331f3daa5be0"
 dependencies = [
  "unicode-ident",
 ]
@@ -4757,6 +4802,15 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "quick-xml"
+version = "0.38.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9845d9dccf565065824e69f9f235fafba1587031eda353c1f1561cd6a6be78f4"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "quinn"
 version = "0.11.8"
@@ -4769,8 +4823,8 @@ dependencies = [
  "quinn-proto",
  "quinn-udp",
  "rustc-hash 2.1.1",
- "rustls 0.23.28",
- "socket2",
+ "rustls 0.23.31",
+ "socket2 0.5.10",
  "thiserror 2.0.12",
  "tokio",
  "tracing",
@@ -4786,10 +4840,10 @@ dependencies = [
  "bytes",
  "getrandom 0.3.3",
  "lru-slab",
- "rand 0.9.1",
+ "rand 0.9.2",
  "ring",
  "rustc-hash 2.1.1",
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "rustls-pki-types",
  "slab",
  "thiserror 2.0.12",
@@ -4807,7 +4861,7 @@ dependencies = [
  "cfg_aliases",
  "libc",
  "once_cell",
- "socket2",
+ "socket2 0.5.10",
  "tracing",
  "windows-sys 0.59.0",
 ]
@@ -4872,9 +4926,9 @@ dependencies = [
 
 [[package]]
 name = "rand"
-version = "0.9.1"
+version = "0.9.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9fbfd9d094a40bf3ae768db9361049ace4c0e04a4fd6b359518bd7b73a73dd97"
+checksum = "6db2770f06117d490610c7488547d543617b21bfa07796d7a12f6f1bd53850d1"
 dependencies = [
  "rand_chacha 0.9.0",
  "rand_core 0.9.3",
@@ -4925,7 +4979,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6a8615d50dcf34fa31f7ab52692afec947c4dd0ab803cc87cb3b0b4570ff7463"
 dependencies = [
  "num-traits",
- "rand 0.9.1",
+ "rand 0.9.2",
 ]
 
 [[package]]
@@ -4958,12 +5012,12 @@ dependencies = [
 
 [[package]]
 name = "rayon-cond"
-version = "0.3.0"
+version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "059f538b55efd2309c9794130bc149c6a553db90e9d99c2030785c82f0bd7df9"
+checksum = "2964d0cf57a3e7a06e8183d14a8b527195c706b7983549cd5462d5aa3747438f"
 dependencies = [
  "either",
- "itertools 0.11.0",
+ "itertools 0.14.0",
  "rayon",
 ]
 
@@ -4985,9 +5039,9 @@ checksum = "03251193000f4bd3b042892be858ee50e8b3719f2b08e5833ac4353724632430"
 
 [[package]]
 name = "redox_syscall"
-version = "0.5.13"
+version = "0.5.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0d04b7d0ee6b4a0207a0a7adb104d23ecb0b47d6beae7152d0fa34b692b29fd6"
+checksum = "5407465600fb0548f1442edf71dd20683c6ed326200ace4b1ef0763521bb3b77"
 dependencies = [
  "bitflags 2.9.1",
 ]
@@ -5005,9 +5059,9 @@ dependencies = [
 
 [[package]]
 name = "redox_users"
-version = "0.5.0"
+version = "0.5.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dd6f9d3d47bdd2ad6945c5015a226ec6155d0bcdfd8f7cd29f86b71f8de99d2b"
+checksum = "a4e608c6638b9c18977b00b475ac1f28d14e84b27d8d42f70e0bf1e3dec127ac"
 dependencies = [
  "getrandom 0.2.16",
  "libredox",
@@ -5100,9 +5154,9 @@ checksum = "2b15c43186be67a4fd63bee50d0303afffcef381492ebe2c5d87f324e1b8815c"
 
 [[package]]
 name = "reqwest"
-version = "0.12.20"
+version = "0.12.22"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "eabf4c97d9130e2bf606614eb937e86edac8292eaa6f422f995d7e8de1eb1813"
+checksum = "cbc931937e6ca3a06e3b6c0aa7841849b160a90351d6ab467a8b9b9959767531"
 dependencies = [
  "async-compression",
  "base64 0.22.1",
@@ -5113,7 +5167,7 @@ dependencies = [
  "futures-channel",
  "futures-core",
  "futures-util",
- "h2 0.4.10",
+ "h2 0.4.12",
  "http 1.3.1",
  "http-body 1.0.1",
  "http-body-util",
@@ -5126,7 +5180,7 @@ dependencies = [
  "percent-encoding",
  "pin-project-lite",
  "quinn",
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "rustls-native-certs 0.8.1",
  "rustls-pki-types",
  "serde",
@@ -5144,7 +5198,7 @@ dependencies = [
  "wasm-bindgen-futures",
  "wasm-streams",
  "web-sys",
- "webpki-roots 1.0.1",
+ "webpki-roots 1.0.2",
 ]
 
 [[package]]
@@ -5197,9 +5251,9 @@ dependencies = [
 
 [[package]]
 name = "rustc-demangle"
-version = "0.1.25"
+version = "0.1.26"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "989e6739f80c4ad5b13e0fd7fe89531180375b18520cc8c82080e4dc4035b84f"
+checksum = "56f7d92ca342cea22a06f2121d944b4fd82af56988c270852495420f961d4ace"
 
 [[package]]
 name = "rustc-hash"
@@ -5237,15 +5291,15 @@ dependencies = [
 
 [[package]]
 name = "rustix"
-version = "1.0.7"
+version = "1.0.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c71e83d6afe7ff64890ec6b71d6a69bb8a610ab78ce364b3352876bb4c801266"
+checksum = "11181fbabf243db407ef8df94a6ce0b2f9a733bd8be4ad02b4eda9602296cac8"
 dependencies = [
  "bitflags 2.9.1",
  "errno",
  "libc",
  "linux-raw-sys 0.9.4",
- "windows-sys 0.59.0",
+ "windows-sys 0.60.2",
 ]
 
 [[package]]
@@ -5262,16 +5316,16 @@ dependencies = [
 
 [[package]]
 name = "rustls"
-version = "0.23.28"
+version = "0.23.31"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7160e3e10bf4535308537f3c4e1641468cd0e485175d6163087c0393c7d46643"
+checksum = "c0ebcbd2f03de0fc1122ad9bb24b127a5a6cd51d72604a3f3c50ac459762b6cc"
 dependencies = [
  "aws-lc-rs",
  "log",
  "once_cell",
  "ring",
  "rustls-pki-types",
- "rustls-webpki 0.103.3",
+ "rustls-webpki 0.103.4",
  "subtle",
  "zeroize",
 ]
@@ -5297,7 +5351,7 @@ dependencies = [
  "openssl-probe",
  "rustls-pki-types",
  "schannel",
- "security-framework 3.2.0",
+ "security-framework 3.3.0",
 ]
 
 [[package]]
@@ -5340,9 +5394,9 @@ dependencies = [
 
 [[package]]
 name = "rustls-webpki"
-version = "0.103.3"
+version = "0.103.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e4a72fe2bcf7a6ac6fd7d0b9e5cb68aeb7d4c0a0271730218b3e92d43b4eb435"
+checksum = "0a17884ae0c1b773f1ccd2bd4a8c72f16da897310a98b0e84bf349ad5ead92fc"
 dependencies = [
  "aws-lc-rs",
  "ring",
@@ -5352,9 +5406,9 @@ dependencies = [
 
 [[package]]
 name = "rustversion"
-version = "1.0.21"
+version = "1.0.22"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8a0d197bd2c9dc6e53b84da9556a69ba4cdfab8619eb41a8bd1cc2027a0f6b1d"
+checksum = "b39cdef0fa800fc44525c84ccb54a029961a8215f9619753635a9c0d2538d46d"
 
 [[package]]
 name = "rustyline"
@@ -5488,9 +5542,9 @@ dependencies = [
 
 [[package]]
 name = "security-framework"
-version = "3.2.0"
+version = "3.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "271720403f46ca04f7ba6f55d438f8bd878d6b8ca0a1046e8228c4145bcbb316"
+checksum = "80fb1d92c5028aa318b4b8bd7302a5bfcf48be96a37fc6fc790f806b0004ee0c"
 dependencies = [
  "bitflags 2.9.1",
  "core-foundation 0.10.1",
@@ -5520,6 +5574,7 @@ dependencies = [
  "candle-transformers",
  "chrono",
  "dirs 5.0.1",
+ "glob",
  "hnsw_rs",
  "indicatif",
  "rayon",
@@ -5593,9 +5648,9 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.140"
+version = "1.0.142"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "20068b6e96dc6c9bd23e01df8827e6c7e1f2fddd43c21810382803c136b99373"
+checksum = "030fedb782600dcbd6f02d479bf0d817ac3bb40d644745b769d6a96bc3afc5a7"
 dependencies = [
  "indexmap",
  "itoa",
@@ -5729,9 +5784,9 @@ dependencies = [
 
 [[package]]
 name = "signal-hook-registry"
-version = "1.4.5"
+version = "1.4.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9203b8055f63a2a00e2f593bb0510367fe707d7ff1e5c872de2f537b339e5410"
+checksum = "b2a4719bff48cee6b39d12c020eeb490953ad2443b7055bd0b21fca26bd8c28b"
 dependencies = [
  "libc",
 ]
@@ -5766,7 +5821,7 @@ dependencies = [
  "indexmap",
  "log",
  "nix 0.29.0",
- "rand 0.9.1",
+ "rand 0.9.2",
  "rayon",
  "regex",
  "shell-quote",
@@ -5781,9 +5836,9 @@ dependencies = [
 
 [[package]]
 name = "slab"
-version = "0.4.10"
+version = "0.4.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "04dc19736151f35336d325007ac991178d504a119863a2fcb3758cdb5e52c50d"
+checksum = "7a2ae44ef20feb57a68b23d846850f861394c2e02dc425a50098ae8c90267589"
 
 [[package]]
 name = "smallvec"
@@ -5801,6 +5856,16 @@ dependencies = [
  "windows-sys 0.52.0",
 ]
 
+[[package]]
+name = "socket2"
+version = "0.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "233504af464074f9d066d7b5416c5f9b894a5862a6506e306f7b816cdd6f1807"
+dependencies = [
+ "libc",
+ "windows-sys 0.59.0",
+]
+
 [[package]]
 name = "spinners"
 version = "4.1.1"
@@ -5830,6 +5895,12 @@ version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3"
 
+[[package]]
+name = "static_assertions"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
+
 [[package]]
 name = "stop-words"
 version = "0.8.1"
@@ -5871,11 +5942,11 @@ checksum = "8fec0f0aef304996cf250b31b5a10dee7980c85da9d759361292b8bca5a18f06"
 
 [[package]]
 name = "strum"
-version = "0.27.1"
+version = "0.27.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f64def088c51c9510a8579e3c5d67c65349dcf755e5479ad3d010aa6454e2c32"
+checksum = "af23d6f6c1a224baef9d3f61e287d2761385a5b88fdab4eb4c6f11aeb54c4bcf"
 dependencies = [
- "strum_macros 0.27.1",
+ "strum_macros 0.27.2",
 ]
 
 [[package]]
@@ -5906,14 +5977,13 @@ dependencies = [
 
 [[package]]
 name = "strum_macros"
-version = "0.27.1"
+version = "0.27.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c77a8c5abcaf0f9ce05d62342b7d298c346515365c36b673df4ebe3ced01fde8"
+checksum = "7695ce3845ea4b33927c055a39dc438a45b059f7c1b3d91d38d10355fb8cbca7"
 dependencies = [
  "heck 0.5.0",
  "proc-macro2",
  "quote",
- "rustversion",
  "syn 2.0.104",
 ]
 
@@ -6068,7 +6138,7 @@ dependencies = [
  "fastrand",
  "getrandom 0.3.3",
  "once_cell",
- "rustix 1.0.7",
+ "rustix 1.0.8",
  "windows-sys 0.59.0",
 ]
 
@@ -6098,7 +6168,7 @@ version = "0.4.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "45c6481c4829e4cc63825e62c49186a34538b7b2750b73b266581ffb612fb5ed"
 dependencies = [
- "rustix 1.0.7",
+ "rustix 1.0.8",
  "windows-sys 0.59.0",
 ]
 
@@ -6246,23 +6316,25 @@ checksum = "1f3ccbac311fea05f86f61904b462b55fb3df8837a366dfc601a0161d0532f20"
 
 [[package]]
 name = "tokenizers"
-version = "0.21.1"
+version = "0.21.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3169b3195f925496c895caee7978a335d49218488ef22375267fba5a46a40bd7"
+checksum = "a620b996116a59e184c2fa2dfd8251ea34a36d0a514758c6f966386bd2e03476"
 dependencies = [
+ "ahash",
  "aho-corasick",
+ "compact_str",
+ "dary_heap",
  "derive_builder",
  "esaxx-rs",
- "getrandom 0.2.16",
+ "getrandom 0.3.3",
  "indicatif",
- "itertools 0.13.0",
- "lazy_static",
+ "itertools 0.14.0",
  "log",
  "macro_rules_attribute",
  "monostate",
  "onig",
  "paste",
- "rand 0.8.5",
+ "rand 0.9.2",
  "rayon",
  "rayon-cond",
  "regex",
@@ -6278,20 +6350,22 @@ dependencies = [
 
 [[package]]
 name = "tokio"
-version = "1.45.1"
+version = "1.47.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "75ef51a33ef1da925cea3e4eb122833cb377c61439ca401b770f54902b806779"
+checksum = "89e49afdadebb872d3145a5638b59eb0691ea23e46ca484037cfab3b76b95038"
 dependencies = [
  "backtrace",
  "bytes",
+ "io-uring",
  "libc",
  "mio",
  "parking_lot",
  "pin-project-lite",
  "signal-hook-registry",
- "socket2",
+ "slab",
+ "socket2 0.6.0",
  "tokio-macros",
- "windows-sys 0.52.0",
+ "windows-sys 0.59.0",
 ]
 
 [[package]]
@@ -6321,7 +6395,7 @@ version = "0.26.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8e727b36a1a0e8b74c376ac2211e40c2c8af09fb4013c60d910495810f008e9b"
 dependencies = [
- "rustls 0.23.28",
+ "rustls 0.23.31",
  "tokio",
 ]
 
@@ -6350,9 +6424,9 @@ dependencies = [
 
 [[package]]
 name = "tokio-util"
-version = "0.7.15"
+version = "0.7.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "66a539a9ad6d5d281510d5bd368c973d636c02dbf8a67300bfb6b950696ad7df"
+checksum = "14307c986784f72ef81c89db7d9e28d6ac26d16213b109ea501696195e6e3ce5"
 dependencies = [
  "bytes",
  "futures-core",
@@ -6394,7 +6468,7 @@ dependencies = [
  "serde_spanned",
  "toml_datetime",
  "toml_write",
- "winnow 0.7.11",
+ "winnow 0.7.12",
 ]
 
 [[package]]
@@ -6570,11 +6644,10 @@ dependencies = [
 
 [[package]]
 name = "tree_magic_mini"
-version = "3.1.6"
+version = "3.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "aac5e8971f245c3389a5a76e648bfc80803ae066a1243a75db0064d7c1129d63"
+checksum = "f943391d896cdfe8eec03a04d7110332d445be7df856db382dd96a730667562c"
 dependencies = [
- "fnv",
  "memchr",
  "nom",
  "once_cell",
@@ -6612,7 +6685,7 @@ dependencies = [
  "http 1.3.1",
  "httparse",
  "log",
- "rand 0.9.1",
+ "rand 0.9.2",
  "sha1",
  "thiserror 2.0.12",
  "utf-8",
@@ -6781,7 +6854,7 @@ checksum = "3cf4199d1e5d15ddd86a694e4d0dffa9c323ce759fea589f00fef9d81cc1931d"
 dependencies = [
  "getrandom 0.3.3",
  "js-sys",
- "rand 0.9.1",
+ "rand 0.9.2",
  "serde",
  "wasm-bindgen",
 ]
@@ -6975,34 +7048,34 @@ dependencies = [
 
 [[package]]
 name = "wayland-backend"
-version = "0.3.10"
+version = "0.3.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fe770181423e5fc79d3e2a7f4410b7799d5aab1de4372853de3c6aa13ca24121"
+checksum = "673a33c33048a5ade91a6b139580fa174e19fb0d23f396dca9fa15f2e1e49b35"
 dependencies = [
  "cc",
  "downcast-rs",
- "rustix 0.38.44",
+ "rustix 1.0.8",
  "smallvec",
  "wayland-sys",
 ]
 
 [[package]]
 name = "wayland-client"
-version = "0.31.10"
+version = "0.31.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "978fa7c67b0847dbd6a9f350ca2569174974cd4082737054dbb7fbb79d7d9a61"
+checksum = "c66a47e840dc20793f2264eb4b3e4ecb4b75d91c0dd4af04b456128e0bdd449d"
 dependencies = [
  "bitflags 2.9.1",
- "rustix 0.38.44",
+ "rustix 1.0.8",
  "wayland-backend",
  "wayland-scanner",
 ]
 
 [[package]]
 name = "wayland-protocols"
-version = "0.32.8"
+version = "0.32.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "779075454e1e9a521794fed15886323ea0feda3f8b0fc1390f5398141310422a"
+checksum = "efa790ed75fbfd71283bd2521a1cfdc022aabcc28bdcff00851f9e4ae88d9901"
 dependencies = [
  "bitflags 2.9.1",
  "wayland-backend",
@@ -7012,9 +7085,9 @@ dependencies = [
 
 [[package]]
 name = "wayland-protocols-wlr"
-version = "0.3.8"
+version = "0.3.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1cb6cdc73399c0e06504c437fe3cf886f25568dd5454473d565085b36d6a8bbf"
+checksum = "efd94963ed43cf9938a090ca4f7da58eb55325ec8200c3848963e98dc25b78ec"
 dependencies = [
  "bitflags 2.9.1",
  "wayland-backend",
@@ -7025,20 +7098,20 @@ dependencies = [
 
 [[package]]
 name = "wayland-scanner"
-version = "0.31.6"
+version = "0.31.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "896fdafd5d28145fce7958917d69f2fd44469b1d4e861cb5961bcbeebc6d1484"
+checksum = "54cb1e9dc49da91950bdfd8b848c49330536d9d1fb03d4bfec8cae50caa50ae3"
 dependencies = [
  "proc-macro2",
- "quick-xml",
+ "quick-xml 0.37.5",
  "quote",
 ]
 
 [[package]]
 name = "wayland-sys"
-version = "0.31.6"
+version = "0.31.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dbcebb399c77d5aa9fa5db874806ee7b4eba4e73650948e8f93963f128896615"
+checksum = "34949b42822155826b41db8e5d0c1be3a2bd296c747577a43a3e6daefc296142"
 dependencies = [
  "pkg-config",
 ]
@@ -7074,9 +7147,9 @@ dependencies = [
 
 [[package]]
 name = "webpki-roots"
-version = "1.0.1"
+version = "1.0.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8782dd5a41a24eed3a4f40b606249b3e236ca61adf1f25ea4d45c73de122b502"
+checksum = "7e8983c3ab33d6fb807cfcdad2491c4ea8cbc8ed839181c7dfd9c67c83e261b2"
 dependencies = [
  "rustls-pki-types",
 ]
@@ -7101,7 +7174,7 @@ checksum = "24d643ce3fd3e5b54854602a080f34fb10ab75e0b813ee32d00ca2b44fa74762"
 dependencies = [
  "either",
  "env_home",
- "rustix 1.0.7",
+ "rustix 1.0.8",
  "winsafe",
 ]
 
@@ -7113,11 +7186,11 @@ checksum = "0b9aa3ad29c3d08283ac6b769e3ec15ad1ddb88af7d2e9bc402c574973b937e7"
 
 [[package]]
 name = "whoami"
-version = "1.6.0"
+version = "1.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6994d13118ab492c3c80c1f81928718159254c53c472bf9ce36f8dae4add02a7"
+checksum = "5d4a4db5077702ca3015d3d02d74974948aba2ad9e12ab7df718ee64ccd7e97d"
 dependencies = [
- "redox_syscall",
+ "libredox",
  "wasite",
  "web-sys",
 ]
@@ -7400,7 +7473,7 @@ version = "0.60.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f2f500e4d28234f72040990ec9d39e3a6b950f9f22d3dba18416c35882612bcb"
 dependencies = [
- "windows-targets 0.53.2",
+ "windows-targets 0.53.3",
 ]
 
 [[package]]
@@ -7436,10 +7509,11 @@ dependencies = [
 
 [[package]]
 name = "windows-targets"
-version = "0.53.2"
+version = "0.53.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c66f69fcc9ce11da9966ddb31a40968cad001c5bedeb5c2b82ede4253ab48aef"
+checksum = "d5fe6031c4041849d7c496a8ded650796e7b6ecc19df1a431c1a363342e5dc91"
 dependencies = [
+ "windows-link",
  "windows_aarch64_gnullvm 0.53.0",
  "windows_aarch64_msvc 0.53.0",
  "windows_i686_gnu 0.53.0",
@@ -7608,9 +7682,9 @@ dependencies = [
 
 [[package]]
 name = "winnow"
-version = "0.7.11"
+version = "0.7.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "74c7b26e3480b707944fc872477815d29a8e429d2f93a1ce000f5fa84a15cbcd"
+checksum = "f3edebf492c8125044983378ecb5766203ad3b4c2f7a922bd7dd207f6d443e95"
 dependencies = [
  "memchr",
 ]
@@ -7811,9 +7885,9 @@ dependencies = [
 
 [[package]]
 name = "zerovec"
-version = "0.11.2"
+version = "0.11.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4a05eb080e015ba39cc9e23bbe5e7fb04d5fb040350f99f34e338d5fdd294428"
+checksum = "e7aa2bd55086f1ab526693ecbe444205da57e25f4489879da80635a46d90e73b"
 dependencies = [
  "yoke 0.8.0",
  "zerofrom",
diff --git a/crates/chat-cli/src/cli/chat/cli/knowledge.rs b/crates/chat-cli/src/cli/chat/cli/knowledge.rs
index 987eda059..7312330ad 100644
--- a/crates/chat-cli/src/cli/chat/cli/knowledge.rs
+++ b/crates/chat-cli/src/cli/chat/cli/knowledge.rs
@@ -29,13 +29,21 @@ pub enum KnowledgeSubcommand {
     /// Display the knowledge base contents
     Show,
     /// Add a file or directory to knowledge base
-    Add { path: String },
-    /// Remove specified knowledge context by path
+    Add {
+        path: String,
+        /// Include patterns (e.g., `**/*.ts`, `**/*.md`)
+        #[arg(long, action = clap::ArgAction::Append)]
+        include: Vec<String>,
+        /// Exclude patterns (e.g., `node_modules/**`, `target/**`)
+        #[arg(long, action = clap::ArgAction::Append)]
+        exclude: Vec<String>,
+    },
+    /// Remove specified knowledge base entry by path
     #[command(alias = "rm")]
     Remove { path: String },
     /// Update a file or directory in knowledge base
     Update { path: String },
-    /// Remove all knowledge contexts
+    /// Remove all knowledge base entries
     Clear,
     /// Show background operation status
     Status,
@@ -79,7 +87,9 @@ impl KnowledgeSubcommand {
         queue!(
             session.stderr,
             style::SetForegroundColor(Color::Red),
-            style::Print("\nKnowledge tool is disabled. Enable it with: q settings chat.enableKnowledge true\n\n"),
+            style::Print("\nKnowledge tool is disabled. Enable it with: q settings chat.enableKnowledge true\n"),
+            style::SetForegroundColor(Color::Yellow),
+            style::Print("💡 Your knowledge base data is preserved and will be available when re-enabled.\n\n"),
             style::SetForegroundColor(Color::Reset)
         )
     }
@@ -93,56 +103,67 @@ impl KnowledgeSubcommand {
     async fn execute_operation(&self, os: &Os, session: &mut ChatSession) -> OperationResult {
         match self {
             KnowledgeSubcommand::Show => {
-                match Self::handle_show(session).await {
+                match Self::handle_show(os, session).await {
                     Ok(_) => OperationResult::Info("".to_string()), // Empty Info, formatting already done
-                    Err(e) => OperationResult::Error(format!("Failed to show contexts: {}", e)),
+                    Err(e) => OperationResult::Error(format!("Failed to show knowledge base entries: {}", e)),
                 }
             },
-            KnowledgeSubcommand::Add { path } => Self::handle_add(os, path).await,
+            KnowledgeSubcommand::Add { path, include, exclude } => Self::handle_add(os, path, include, exclude).await,
             KnowledgeSubcommand::Remove { path } => Self::handle_remove(os, path).await,
             KnowledgeSubcommand::Update { path } => Self::handle_update(os, path).await,
-            KnowledgeSubcommand::Clear => Self::handle_clear(session).await,
-            KnowledgeSubcommand::Status => Self::handle_status().await,
-            KnowledgeSubcommand::Cancel { operation_id } => Self::handle_cancel(operation_id.as_deref()).await,
+            KnowledgeSubcommand::Clear => Self::handle_clear(os, session).await,
+            KnowledgeSubcommand::Status => Self::handle_status(os).await,
+            KnowledgeSubcommand::Cancel { operation_id } => Self::handle_cancel(os, operation_id.as_deref()).await,
         }
     }
 
-    async fn handle_show(session: &mut ChatSession) -> Result<(), std::io::Error> {
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
-        let store = async_knowledge_store.lock().await;
-
-        // Use the async get_all method which is concurrent with indexing
-        let contexts = store.get_all().await.unwrap_or_else(|e| {
-            // Write error to output using queue system
-            let _ = queue!(
-                session.stderr,
-                style::SetForegroundColor(Color::Red),
-                style::Print(&format!("Error getting contexts: {}\n", e)),
-                style::ResetColor
-            );
-            Vec::new()
-        });
-
-        Self::format_contexts(session, &contexts)
+    async fn handle_show(os: &Os, session: &mut ChatSession) -> Result<(), std::io::Error> {
+        match KnowledgeStore::get_async_instance_with_os(os).await {
+            Ok(store) => {
+                let store = store.lock().await;
+                let entries = store.get_all().await.unwrap_or_else(|e| {
+                    let _ = queue!(
+                        session.stderr,
+                        style::SetForegroundColor(Color::Red),
+                        style::Print(&format!("Error getting knowledge base entries: {}\n", e)),
+                        style::ResetColor
+                    );
+                    Vec::new()
+                });
+                let _ = Self::format_knowledge_entries(session, &entries);
+            },
+            Err(e) => {
+                queue!(
+                    session.stderr,
+                    style::SetForegroundColor(Color::Red),
+                    style::Print(&format!("Error accessing knowledge base: {}\n", e)),
+                    style::SetForegroundColor(Color::Reset)
+                )?;
+            },
+        }
+        Ok(())
     }
 
-    fn format_contexts(session: &mut ChatSession, contexts: &[KnowledgeContext]) -> Result<(), std::io::Error> {
-        if contexts.is_empty() {
+    fn format_knowledge_entries(
+        session: &mut ChatSession,
+        knowledge_entries: &[KnowledgeContext],
+    ) -> Result<(), std::io::Error> {
+        if knowledge_entries.is_empty() {
             queue!(
                 session.stderr,
                 style::Print("\nNo knowledge base entries found.\n"),
-                style::Print("💡 Tip: If indexing is in progress, contexts may not appear until indexing completes.\n"),
+                style::Print("💡 Tip: If indexing is in progress, entries may not appear until indexing completes.\n"),
                 style::Print("   Use 'knowledge status' to check active operations.\n\n")
             )?;
         } else {
             queue!(
                 session.stderr,
-                style::Print("\n📚 Knowledge Base Contexts:\n"),
+                style::Print("\n📚 Knowledge Base Entries:\n"),
                 style::Print(format!("{}\n", "━".repeat(80)))
             )?;
 
-            for context in contexts {
-                Self::format_single_context(session, &context)?;
+            for entry in knowledge_entries {
+                Self::format_single_entry(session, &entry)?;
                 queue!(session.stderr, style::Print(format!("{}\n", "━".repeat(80))))?;
             }
             // Add final newline to match original formatting exactly
@@ -151,32 +172,32 @@ impl KnowledgeSubcommand {
         Ok(())
     }
 
-    fn format_single_context(session: &mut ChatSession, context: &&KnowledgeContext) -> Result<(), std::io::Error> {
+    fn format_single_entry(session: &mut ChatSession, entry: &&KnowledgeContext) -> Result<(), std::io::Error> {
         queue!(
             session.stderr,
             style::SetAttribute(style::Attribute::Bold),
             style::SetForegroundColor(Color::Cyan),
-            style::Print(format!("📂 {}: ", context.id)),
+            style::Print(format!("📂 {}: ", entry.id)),
             style::SetForegroundColor(Color::Green),
-            style::Print(&context.name),
+            style::Print(&entry.name),
             style::SetAttribute(style::Attribute::Reset),
             style::Print("\n")
         )?;
 
         queue!(
             session.stderr,
-            style::Print(format!("   Description: {}\n", context.description)),
+            style::Print(format!("   Description: {}\n", entry.description)),
             style::Print(format!(
                 "   Created: {}\n",
-                context.created_at.format("%Y-%m-%d %H:%M:%S")
+                entry.created_at.format("%Y-%m-%d %H:%M:%S")
             )),
             style::Print(format!(
                 "   Updated: {}\n",
-                context.updated_at.format("%Y-%m-%d %H:%M:%S")
+                entry.updated_at.format("%Y-%m-%d %H:%M:%S")
             ))
         )?;
 
-        if let Some(path) = &context.source_path {
+        if let Some(path) = &entry.source_path {
             queue!(session.stderr, style::Print(format!("   Source: {}\n", path)))?;
         }
 
@@ -184,12 +205,12 @@ impl KnowledgeSubcommand {
             session.stderr,
             style::Print("   Items: "),
             style::SetForegroundColor(Color::Yellow),
-            style::Print(format!("{}", context.item_count)),
+            style::Print(format!("{}", entry.item_count)),
             style::SetForegroundColor(Color::Reset),
             style::Print(" | Persistent: ")
         )?;
 
-        if context.persistent {
+        if entry.persistent {
             queue!(
                 session.stderr,
                 style::SetForegroundColor(Color::Green),
@@ -210,33 +231,58 @@ impl KnowledgeSubcommand {
     }
 
     /// Handle add operation
-    async fn handle_add(os: &Os, path: &str) -> OperationResult {
+    async fn handle_add(
+        os: &Os,
+        path: &str,
+        include_patterns: &[String],
+        exclude_patterns: &[String],
+    ) -> OperationResult {
         match Self::validate_and_sanitize_path(os, path) {
             Ok(sanitized_path) => {
-                let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+                let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+                    Ok(store) => store,
+                    Err(e) => return OperationResult::Error(format!("Error accessing knowledge base: {}", e)),
+                };
                 let mut store = async_knowledge_store.lock().await;
 
-                // Use the async add method which is fire-and-forget
-                match store.add(path, &sanitized_path).await {
+                let options = if include_patterns.is_empty() && exclude_patterns.is_empty() {
+                    crate::util::knowledge_store::AddOptions::with_db_defaults(os)
+                } else {
+                    crate::util::knowledge_store::AddOptions::new()
+                        .with_include_patterns(include_patterns.to_vec())
+                        .with_exclude_patterns(exclude_patterns.to_vec())
+                };
+
+                match store.add(path, &sanitized_path.clone(), options).await {
                     Ok(message) => OperationResult::Info(message),
-                    Err(e) => OperationResult::Error(format!("Failed to add to knowledge base: {}", e)),
+                    Err(e) => {
+                        if e.contains("Invalid include pattern") || e.contains("Invalid exclude pattern") {
+                            OperationResult::Error(e)
+                        } else {
+                            OperationResult::Error(format!("Failed to add: {}", e))
+                        }
+                    },
                 }
             },
-            Err(e) => OperationResult::Error(e),
+            Err(e) => OperationResult::Error(format!("Invalid path: {}", e)),
         }
     }
 
     /// Handle remove operation
     async fn handle_remove(os: &Os, path: &str) -> OperationResult {
         let sanitized_path = sanitize_path_tool_arg(os, path);
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+
+        let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+            Ok(store) => store,
+            Err(e) => return OperationResult::Error(format!("Error accessing knowledge base: {}", e)),
+        };
         let mut store = async_knowledge_store.lock().await;
 
         // Try path first, then name
         if store.remove_by_path(&sanitized_path.to_string_lossy()).await.is_ok() {
-            OperationResult::Success(format!("Removed context with path '{}'", path))
+            OperationResult::Success(format!("Removed knowledge base entry with path '{}'", path))
         } else if store.remove_by_name(path).await.is_ok() {
-            OperationResult::Success(format!("Removed context with name '{}'", path))
+            OperationResult::Success(format!("Removed knowledge base entry with name '{}'", path))
         } else {
             OperationResult::Warning(format!("Entry not found in knowledge base: {}", path))
         }
@@ -246,7 +292,12 @@ impl KnowledgeSubcommand {
     async fn handle_update(os: &Os, path: &str) -> OperationResult {
         match Self::validate_and_sanitize_path(os, path) {
             Ok(sanitized_path) => {
-                let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+                let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+                    Ok(store) => store,
+                    Err(e) => {
+                        return OperationResult::Error(format!("Error accessing knowledge base directory: {}", e));
+                    },
+                };
                 let mut store = async_knowledge_store.lock().await;
 
                 match store.update_by_path(&sanitized_path).await {
@@ -259,11 +310,12 @@ impl KnowledgeSubcommand {
     }
 
     /// Handle clear operation
-    async fn handle_clear(session: &mut ChatSession) -> OperationResult {
+    async fn handle_clear(os: &Os, session: &mut ChatSession) -> OperationResult {
         // Require confirmation
         queue!(
             session.stderr,
-            style::Print("⚠️  This will remove ALL knowledge base entries. Are you sure? (y/N): ")
+            style::Print("⚠️  This action will remove all knowledge base entries.\n"),
+            style::Print("Clear the knowledge base? (y/N): ")
         )
         .unwrap();
         session.stderr.flush().unwrap();
@@ -278,7 +330,10 @@ impl KnowledgeSubcommand {
             return OperationResult::Info("Clear operation cancelled".to_string());
         }
 
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+        let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+            Ok(store) => store,
+            Err(e) => return OperationResult::Error(format!("Error accessing knowledge base directory: {}", e)),
+        };
         let mut store = async_knowledge_store.lock().await;
 
         // First, cancel any pending operations
@@ -308,8 +363,11 @@ impl KnowledgeSubcommand {
     }
 
     /// Handle status operation
-    async fn handle_status() -> OperationResult {
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+    async fn handle_status(os: &Os) -> OperationResult {
+        let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+            Ok(store) => store,
+            Err(e) => return OperationResult::Error(format!("Error accessing knowledge base directory: {}", e)),
+        };
         let store = async_knowledge_store.lock().await;
 
         match store.get_status_data().await {
@@ -325,9 +383,9 @@ impl KnowledgeSubcommand {
     fn format_status_display(status: &SystemStatus) -> String {
         let mut status_lines = Vec::new();
 
-        // Show context summary
+        // Show knowledge base summary
         status_lines.push(format!(
-            "📚 Total contexts: {} ({} persistent, {} volatile)",
+            "📚 Total knowledge base entries: {} ({} persistent, {} volatile)",
             status.total_contexts, status.persistent_contexts, status.volatile_contexts
         ));
 
@@ -416,8 +474,11 @@ impl KnowledgeSubcommand {
     }
 
     /// Handle cancel operation
-    async fn handle_cancel(operation_id: Option<&str>) -> OperationResult {
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+    async fn handle_cancel(os: &Os, operation_id: Option<&str>) -> OperationResult {
+        let async_knowledge_store = match KnowledgeStore::get_async_instance_with_os(os).await {
+            Ok(store) => store,
+            Err(e) => return OperationResult::Error(format!("Error accessing knowledge base directory: {}", e)),
+        };
         let mut store = async_knowledge_store.lock().await;
 
         match store.cancel_operation(operation_id).await {
@@ -491,3 +552,125 @@ impl KnowledgeSubcommand {
         }
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use clap::Parser;
+
+    use super::*;
+
+    #[derive(Parser)]
+    #[command(name = "test")]
+    struct TestCli {
+        #[command(subcommand)]
+        knowledge: KnowledgeSubcommand,
+    }
+
+    #[test]
+    fn test_include_exclude_patterns_parsing() {
+        // Test that include and exclude patterns are parsed correctly
+        let result = TestCli::try_parse_from(&[
+            "test",
+            "add",
+            "/some/path",
+            "--include",
+            "*.rs",
+            "--include",
+            "**/*.md",
+            "--exclude",
+            "node_modules/**",
+            "--exclude",
+            "target/**",
+        ]);
+
+        assert!(result.is_ok());
+        let cli = result.unwrap();
+
+        if let KnowledgeSubcommand::Add { path, include, exclude } = cli.knowledge {
+            assert_eq!(path, "/some/path");
+            assert_eq!(include, vec!["*.rs", "**/*.md"]);
+            assert_eq!(exclude, vec!["node_modules/**", "target/**"]);
+        } else {
+            panic!("Expected Add subcommand");
+        }
+    }
+
+    #[test]
+    fn test_clap_markdown_parsing_issue() {
+        let help_result = TestCli::try_parse_from(&["test", "add", "--help"]);
+        match help_result {
+            Err(err) if err.kind() == clap::error::ErrorKind::DisplayHelp => {
+                // This is expected for --help
+                // The actual issue would be visible in the help text formatting
+                // We can't easily test the exact formatting here, but this documents the issue
+            },
+            _ => panic!("Expected help output"),
+        }
+    }
+
+    #[test]
+    fn test_empty_patterns_allowed() {
+        // Test that commands work without any patterns
+        let result = TestCli::try_parse_from(&["test", "add", "/some/path"]);
+        assert!(result.is_ok());
+
+        let cli = result.unwrap();
+        if let KnowledgeSubcommand::Add { path, include, exclude } = cli.knowledge {
+            assert_eq!(path, "/some/path");
+            assert!(include.is_empty());
+            assert!(exclude.is_empty());
+        } else {
+            panic!("Expected Add subcommand");
+        }
+    }
+
+    #[test]
+    fn test_multiple_include_patterns() {
+        // Test multiple include patterns
+        let result = TestCli::try_parse_from(&[
+            "test",
+            "add",
+            "/some/path",
+            "--include",
+            "*.rs",
+            "--include",
+            "*.md",
+            "--include",
+            "*.txt",
+        ]);
+
+        assert!(result.is_ok());
+        let cli = result.unwrap();
+
+        if let KnowledgeSubcommand::Add { include, .. } = cli.knowledge {
+            assert_eq!(include, vec!["*.rs", "*.md", "*.txt"]);
+        } else {
+            panic!("Expected Add subcommand");
+        }
+    }
+
+    #[test]
+    fn test_multiple_exclude_patterns() {
+        // Test multiple exclude patterns
+        let result = TestCli::try_parse_from(&[
+            "test",
+            "add",
+            "/some/path",
+            "--exclude",
+            "node_modules/**",
+            "--exclude",
+            "target/**",
+            "--exclude",
+            ".git/**",
+        ]);
+
+        assert!(result.is_ok());
+        let cli = result.unwrap();
+
+        if let KnowledgeSubcommand::Add { exclude, .. } = cli.knowledge {
+            assert_eq!(exclude, vec!["node_modules/**", "target/**", ".git/**"]);
+        } else {
+            panic!("Expected Add subcommand");
+        }
+    }
+}
diff --git a/crates/chat-cli/src/cli/chat/tools/knowledge.rs b/crates/chat-cli/src/cli/chat/tools/knowledge.rs
index 7e9bcef21..191fbb86d 100644
--- a/crates/chat-cli/src/cli/chat/tools/knowledge.rs
+++ b/crates/chat-cli/src/cli/chat/tools/knowledge.rs
@@ -124,7 +124,9 @@ impl Knowledge {
             Knowledge::Update(update) => {
                 // Require at least one identifier (context_id or name)
                 if update.context_id.is_empty() && update.name.is_empty() && update.path.is_empty() {
-                    eyre::bail!("Please provide either context_id or name or path to identify the context to update");
+                    eyre::bail!(
+                        "Please provide either context_id, name, or path to identify the knowledge base entry to update"
+                    );
                 }
 
                 // Validate the path exists
@@ -310,8 +312,10 @@ impl Knowledge {
     }
 
     pub async fn invoke(&self, os: &Os, _updates: &mut impl Write) -> Result<InvokeOutput> {
-        // Get the async knowledge store singleton
-        let async_knowledge_store = KnowledgeStore::get_async_instance().await;
+        // Get the async knowledge store singleton with OS-aware directory
+        let async_knowledge_store = KnowledgeStore::get_async_instance_with_os(os)
+            .await
+            .map_err(|e| eyre::eyre!("Failed to access knowledge base: {}", e))?;
         let mut store = async_knowledge_store.lock().await;
 
         let result = match self {
@@ -325,7 +329,14 @@ impl Knowledge {
                     add.value.clone()
                 };
 
-                match store.add(&add.name, &value_to_use).await {
+                match store
+                    .add(
+                        &add.name,
+                        &value_to_use,
+                        crate::util::knowledge_store::AddOptions::with_db_defaults(os),
+                    )
+                    .await
+                {
                     Ok(context_id) => format!(
                         "Added '{}' to knowledge base with ID: {}. Track active jobs in '/knowledge status' with provided id.",
                         add.name, context_id
@@ -412,23 +423,24 @@ impl Knowledge {
                 .await
                 .unwrap_or_else(|e| format!("Failed to clear knowledge base: {}", e)),
             Knowledge::Search(search) => {
-                // Only use a spinner for search, not a full progress bar
                 let results = store.search(&search.query, search.context_id.as_deref()).await;
                 match results {
                     Ok(results) => {
                         if results.is_empty() {
-                            "No matching entries found in knowledge base".to_string()
+                            format!("No matching entries found for query: \"{}\"", search.query)
                         } else {
-                            let mut output = String::from("Search results:\n");
+                            let mut output = format!("Search results for \"{}\":\n\n", search.query);
                             for result in results {
                                 if let Some(text) = result.text() {
-                                    output.push_str(&format!("- {}\n", text));
+                                    output.push_str(&format!("{}\n\n", text));
                                 }
                             }
                             output
                         }
                     },
-                    Err(e) => format!("Search failed: {}", e),
+                    Err(e) => {
+                        format!("Search failed: {}", e)
+                    },
                 }
             },
             Knowledge::Show => {
diff --git a/crates/chat-cli/src/database/settings.rs b/crates/chat-cli/src/database/settings.rs
index 8cfba1f07..519ac445b 100644
--- a/crates/chat-cli/src/database/settings.rs
+++ b/crates/chat-cli/src/database/settings.rs
@@ -22,6 +22,11 @@ pub enum Setting {
     ShareCodeWhispererContent,
     EnabledThinking,
     EnabledKnowledge,
+    KnowledgeDefaultIncludePatterns,
+    KnowledgeDefaultExcludePatterns,
+    KnowledgeMaxFiles,
+    KnowledgeChunkSize,
+    KnowledgeChunkOverlap,
     SkimCommandKey,
     ChatGreetingEnabled,
     ApiTimeout,
@@ -47,6 +52,11 @@ impl AsRef<str> for Setting {
             Self::ShareCodeWhispererContent => "codeWhisperer.shareCodeWhispererContentWithAWS",
             Self::EnabledThinking => "chat.enableThinking",
             Self::EnabledKnowledge => "chat.enableKnowledge",
+            Self::KnowledgeDefaultIncludePatterns => "knowledge.defaultIncludePatterns",
+            Self::KnowledgeDefaultExcludePatterns => "knowledge.defaultExcludePatterns",
+            Self::KnowledgeMaxFiles => "knowledge.maxFiles",
+            Self::KnowledgeChunkSize => "knowledge.chunkSize",
+            Self::KnowledgeChunkOverlap => "knowledge.chunkOverlap",
             Self::SkimCommandKey => "chat.skimCommandKey",
             Self::ChatGreetingEnabled => "chat.greeting.enabled",
             Self::ApiTimeout => "api.timeout",
@@ -82,6 +92,11 @@ impl TryFrom<&str> for Setting {
             "codeWhisperer.shareCodeWhispererContentWithAWS" => Ok(Self::ShareCodeWhispererContent),
             "chat.enableThinking" => Ok(Self::EnabledThinking),
             "chat.enableKnowledge" => Ok(Self::EnabledKnowledge),
+            "knowledge.defaultIncludePatterns" => Ok(Self::KnowledgeDefaultIncludePatterns),
+            "knowledge.defaultExcludePatterns" => Ok(Self::KnowledgeDefaultExcludePatterns),
+            "knowledge.maxFiles" => Ok(Self::KnowledgeMaxFiles),
+            "knowledge.chunkSize" => Ok(Self::KnowledgeChunkSize),
+            "knowledge.chunkOverlap" => Ok(Self::KnowledgeChunkOverlap),
             "chat.skimCommandKey" => Ok(Self::SkimCommandKey),
             "chat.greeting.enabled" => Ok(Self::ChatGreetingEnabled),
             "api.timeout" => Ok(Self::ApiTimeout),
@@ -166,6 +181,10 @@ impl Settings {
         self.get(key).and_then(|value| value.as_i64())
     }
 
+    pub fn get_int_or(&self, key: Setting, default: usize) -> usize {
+        self.get_int(key).map_or(default, |v| v as usize)
+    }
+
     pub async fn save_to_file(&self) -> Result<(), DatabaseError> {
         if cfg!(test) {
             return Ok(());
diff --git a/crates/chat-cli/src/util/directories.rs b/crates/chat-cli/src/util/directories.rs
index d5fb2ed04..a726cf376 100644
--- a/crates/chat-cli/src/util/directories.rs
+++ b/crates/chat-cli/src/util/directories.rs
@@ -185,6 +185,11 @@ pub fn chat_profiles_dir(os: &Os) -> Result<PathBuf> {
     Ok(home_dir(os)?.join(".aws").join("amazonq").join("profiles"))
 }
 
+/// The directory for knowledge base storage
+pub fn knowledge_bases_dir(os: &Os) -> Result<PathBuf> {
+    Ok(home_dir(os)?.join(".aws").join("amazonq").join("knowledge_bases"))
+}
+
 /// The path to the fig settings file
 pub fn settings_path() -> Result<PathBuf> {
     Ok(fig_data_dir()?.join("settings.json"))
diff --git a/crates/chat-cli/src/util/knowledge_store.rs b/crates/chat-cli/src/util/knowledge_store.rs
index 23ef1345e..92370f7a8 100644
--- a/crates/chat-cli/src/util/knowledge_store.rs
+++ b/crates/chat-cli/src/util/knowledge_store.rs
@@ -1,3 +1,4 @@
+use std::path::PathBuf;
 use std::sync::{
     Arc,
     LazyLock as Lazy,
@@ -6,10 +7,74 @@ use std::sync::{
 use eyre::Result;
 use semantic_search_client::KnowledgeContext;
 use semantic_search_client::client::AsyncSemanticSearchClient;
-use semantic_search_client::types::SearchResult;
+use semantic_search_client::types::{
+    AddContextRequest,
+    SearchResult,
+};
 use tokio::sync::Mutex;
+use tracing::debug;
 use uuid::Uuid;
 
+use crate::os::Os;
+use crate::util::directories;
+
+/// Configuration for adding knowledge contexts
+#[derive(Default)]
+pub struct AddOptions {
+    pub description: Option<String>,
+    pub include_patterns: Vec<String>,
+    pub exclude_patterns: Vec<String>,
+}
+
+impl AddOptions {
+    pub fn new() -> Self {
+        Self::default()
+    }
+
+    /// Create AddOptions with DB default patterns
+    pub fn with_db_defaults(os: &crate::os::Os) -> Self {
+        let default_include = os
+            .database
+            .settings
+            .get(crate::database::settings::Setting::KnowledgeDefaultIncludePatterns)
+            .and_then(|v| v.as_array())
+            .map(|arr| {
+                arr.iter()
+                    .filter_map(|v| v.as_str().map(|s| s.to_string()))
+                    .collect::<Vec<_>>()
+            })
+            .unwrap_or_default();
+
+        let default_exclude = os
+            .database
+            .settings
+            .get(crate::database::settings::Setting::KnowledgeDefaultExcludePatterns)
+            .and_then(|v| v.as_array())
+            .map(|arr| {
+                arr.iter()
+                    .filter_map(|v| v.as_str().map(|s| s.to_string()))
+                    .collect::<Vec<_>>()
+            })
+            .unwrap_or_default();
+
+        Self {
+            description: None,
+            include_patterns: default_include,
+            exclude_patterns: default_exclude,
+        }
+    }
+
+    pub fn with_include_patterns(mut self, patterns: Vec<String>) -> Self {
+        self.include_patterns = patterns;
+        self
+    }
+
+    pub fn with_exclude_patterns(mut self, patterns: Vec<String>) -> Self {
+        self.exclude_patterns = patterns;
+        self
+    }
+}
+
 #[derive(Debug)]
 pub enum KnowledgeError {
     ClientError(String),
@@ -31,14 +96,57 @@ pub struct KnowledgeStore {
 }
 
 impl KnowledgeStore {
-    /// Get singleton instance
-    pub async fn get_async_instance() -> Arc<Mutex<Self>> {
+    /// Get singleton instance with directory from OS (includes migration)
+    pub async fn get_async_instance_with_os(os: &Os) -> Result<Arc<Mutex<Self>>, directories::DirectoryError> {
+        let knowledge_dir = crate::util::directories::knowledge_bases_dir(os)?;
+        Self::migrate_legacy_knowledge_base(&knowledge_dir).await;
+        Ok(Self::get_async_instance_with_os_settings(os, knowledge_dir).await)
+    }
+
+    /// Migrate legacy knowledge base from old location if needed
+    async fn migrate_legacy_knowledge_base(knowledge_dir: &PathBuf) {
+        let old_flat_dir = dirs::home_dir()
+            .unwrap_or_else(|| PathBuf::from("."))
+            .join(".semantic_search");
+
+        if old_flat_dir.exists() && !knowledge_dir.exists() {
+            // Create parent directories first
+            if let Some(parent) = knowledge_dir.parent() {
+                if let Err(e) = std::fs::create_dir_all(parent) {
+                    debug!(
+                        "Warning: Failed to create parent directories for knowledge base migration: {}",
+                        e
+                    );
+                    return;
+                }
+            }
+
+            // Attempt migration
+            if let Err(e) = std::fs::rename(&old_flat_dir, knowledge_dir) {
+                debug!(
+                    "Warning: Failed to migrate legacy knowledge base from {} to {}: {}",
+                    old_flat_dir.display(),
+                    knowledge_dir.display(),
+                    e
+                );
+            } else {
+                println!(
+                    "✅ Migrated knowledge base from {} to {}",
+                    old_flat_dir.display(),
+                    knowledge_dir.display()
+                );
+            }
+        }
+    }
+
+    /// Get singleton instance with OS settings (primary method)
+    pub async fn get_async_instance_with_os_settings(os: &crate::os::Os, base_dir: PathBuf) -> Arc<Mutex<Self>> {
         static ASYNC_INSTANCE: Lazy<tokio::sync::OnceCell<Arc<Mutex<KnowledgeStore>>>> =
             Lazy::new(tokio::sync::OnceCell::new);
 
         if cfg!(test) {
             Arc::new(Mutex::new(
-                KnowledgeStore::new()
+                KnowledgeStore::new_with_os_settings(os, base_dir)
                     .await
                     .expect("Failed to create test async knowledge store"),
             ))
@@ -46,7 +154,7 @@ impl KnowledgeStore {
             ASYNC_INSTANCE
                 .get_or_init(|| async {
                     Arc::new(Mutex::new(
-                        KnowledgeStore::new()
+                        KnowledgeStore::new_with_os_settings(os, base_dir)
                             .await
                             .expect("Failed to create async knowledge store"),
                     ))
@@ -56,37 +164,126 @@ impl KnowledgeStore {
         }
     }
 
-    pub async fn new() -> Result<Self> {
-        let client = AsyncSemanticSearchClient::new_with_default_dir()
+    /// Create SemanticSearchConfig from database settings with fallbacks to defaults
+    fn create_config_from_db_settings(
+        os: &crate::os::Os,
+        base_dir: PathBuf,
+    ) -> semantic_search_client::config::SemanticSearchConfig {
+        use semantic_search_client::config::SemanticSearchConfig;
+
+        use crate::database::settings::Setting;
+
+        // Create default config first
+        let default_config = SemanticSearchConfig {
+            base_dir: base_dir.clone(),
+            ..Default::default()
+        };
+
+        // Override with DB settings if provided, otherwise use defaults
+        let chunk_size = os
+            .database
+            .settings
+            .get_int_or(Setting::KnowledgeChunkSize, default_config.chunk_size);
+        let chunk_overlap = os
+            .database
+            .settings
+            .get_int_or(Setting::KnowledgeChunkOverlap, default_config.chunk_overlap);
+        let max_files = os
+            .database
+            .settings
+            .get_int_or(Setting::KnowledgeMaxFiles, default_config.max_files);
+
+        SemanticSearchConfig {
+            chunk_size,
+            chunk_overlap,
+            max_files,
+            base_dir,
+            ..default_config
+        }
+    }
+
+    /// Create instance with database settings from OS
+    pub async fn new_with_os_settings(os: &crate::os::Os, base_dir: PathBuf) -> Result<Self> {
+        let config = Self::create_config_from_db_settings(os, base_dir.clone());
+        let client = AsyncSemanticSearchClient::with_config(&base_dir, config)
             .await
             .map_err(|e| eyre::eyre!("Failed to create client: {}", e))?;
 
         Ok(Self { client })
     }
 
-    /// Add context - delegates to async client
-    pub async fn add(&mut self, name: &str, path_str: &str) -> Result<String, String> {
+    /// Add context with flexible options
+    pub async fn add(&mut self, name: &str, path_str: &str, options: AddOptions) -> Result<String, String> {
         let path_buf = std::path::PathBuf::from(path_str);
         let canonical_path = path_buf
             .canonicalize()
             .map_err(|_io_error| format!("❌ Path does not exist: {}", path_str))?;
 
-        match self
-            .client
-            .add_context_from_path(&canonical_path, name, &format!("Knowledge context for {}", name), true)
-            .await
-        {
-            Ok((operation_id, _)) => Ok(format!(
-                "🚀 Started indexing '{}'\n📁 Path: {}\n🆔 Operation ID: {}.",
-                name,
-                canonical_path.display(),
-                &operation_id.to_string()[..8]
-            )),
-            Err(e) => Err(format!("Failed to start indexing: {}", e)),
+        // Use provided description or generate default
+        let description = options
+            .description
+            .unwrap_or_else(|| format!("Knowledge context for {}", name));
+
+        // Create AddContextRequest with all options
+        let request = AddContextRequest {
+            path: canonical_path.clone(),
+            name: name.to_string(),
+            description: if !options.include_patterns.is_empty() || !options.exclude_patterns.is_empty() {
+                let mut full_description = description;
+                if !options.include_patterns.is_empty() {
+                    full_description.push_str(&format!(" [Include: {}]", options.include_patterns.join(", ")));
+                }
+                if !options.exclude_patterns.is_empty() {
+                    full_description.push_str(&format!(" [Exclude: {}]", options.exclude_patterns.join(", ")));
+                }
+                full_description
+            } else {
+                description
+            },
+            persistent: true,
+            include_patterns: if options.include_patterns.is_empty() {
+                None
+            } else {
+                Some(options.include_patterns.clone())
+            },
+            exclude_patterns: if options.exclude_patterns.is_empty() {
+                None
+            } else {
+                Some(options.exclude_patterns.clone())
+            },
+        };
+
+        match self.client.add_context(request).await {
+            Ok((operation_id, _)) => {
+                let mut message = format!(
+                    "🚀 Started indexing '{}'\n📁 Path: {}\n🆔 Operation ID: {}",
+                    name,
+                    canonical_path.display(),
+                    &operation_id.to_string()[..8]
+                );
+                if !options.include_patterns.is_empty() || !options.exclude_patterns.is_empty() {
+                    message.push_str("\n📋 Pattern filtering applied:");
+                    if !options.include_patterns.is_empty() {
+                        message.push_str(&format!("\n   Include: {}", options.include_patterns.join(", ")));
+                    }
+                    if !options.exclude_patterns.is_empty() {
+                        message.push_str(&format!("\n   Exclude: {}", options.exclude_patterns.join(", ")));
+                    }
+                    message.push_str("\n✅ Only matching files will be indexed");
+                }
+                Ok(message)
+            },
+            Err(e) => {
+                let error_msg = e.to_string();
+                if error_msg.contains("Invalid include pattern") || error_msg.contains("Invalid exclude pattern") {
+                    Err(error_msg)
+                } else {
+                    Err(format!("Failed to start indexing: {}", e))
+                }
+            },
         }
     }
 
-    /// Get all contexts - delegates to async client
     pub async fn get_all(&self) -> Result<Vec<KnowledgeContext>, KnowledgeError> {
         Ok(self.client.get_contexts().await)
     }
@@ -142,8 +339,11 @@ impl KnowledgeStore {
                 }
             }
         } else {
-            // Cancel all operations
-            self.client.cancel_all_operations().await.map_err(|e| e.to_string())
+            // Cancel most recent operation (not all operations)
+            self.client
+                .cancel_most_recent_operation()
+                .await
+                .map_err(|e| e.to_string())
         }
     }
 
@@ -207,8 +407,13 @@ impl KnowledgeStore {
                 .await
                 .map_err(|e| e.to_string())?;
 
-            // Then add it back with the same name
-            self.add(&context.name, path_str).await
+            // Then add it back with the same name and original patterns
+            let options = AddOptions {
+                description: None,
+                include_patterns: context.include_patterns.clone(),
+                exclude_patterns: context.exclude_patterns.clone(),
+            };
+            self.add(&context.name, path_str, options).await
         } else {
             // Debug: List all available contexts
             let available_paths = self.client.list_context_paths().await;
@@ -240,8 +445,13 @@ impl KnowledgeStore {
             .await
             .map_err(|e| e.to_string())?;
 
-        // Then add it back with the same name
-        self.add(&context_name, path_str).await
+        // Then add it back with the same name and original patterns
+        let options = AddOptions {
+            description: None,
+            include_patterns: context.include_patterns.clone(),
+            exclude_patterns: context.exclude_patterns.clone(),
+        };
+        self.add(&context_name, path_str, options).await
     }
 
     /// Update context by name
@@ -253,10 +463,59 @@ impl KnowledgeStore {
                 .await
                 .map_err(|e| e.to_string())?;
 
-            // Then add it back with the same name
-            self.add(name, path_str).await
+            // Then add it back with the same name and original patterns
+            let options = AddOptions {
+                description: None,
+                include_patterns: context.include_patterns.clone(),
+                exclude_patterns: context.exclude_patterns.clone(),
+            };
+            self.add(name, path_str, options).await
         } else {
             Err(format!("Context with name '{}' not found", name))
         }
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use tempfile::TempDir;
+
+    use super::*;
+    use crate::os::Os;
+
+    async fn create_test_os(temp_dir: &TempDir) -> Os {
+        let os = Os::new().await.unwrap();
+        // Override home directory to use temp directory
+        unsafe {
+            os.env.set_var("HOME", temp_dir.path().to_str().unwrap());
+        }
+        os
+    }
+
+    #[tokio::test]
+    async fn test_create_config_from_db_settings() {
+        let temp_dir = TempDir::new().unwrap();
+        let os = create_test_os(&temp_dir).await;
+        let base_dir = temp_dir.path().join("test_kb");
+
+        // Test config creation with default settings
+        let config = KnowledgeStore::create_config_from_db_settings(&os, base_dir.clone());
+
+        // Should use defaults when no database settings exist
+        assert_eq!(config.chunk_size, 512); // Default chunk size
+        assert_eq!(config.chunk_overlap, 128); // Default chunk overlap
+        assert_eq!(config.max_files, 10000); // Default max files
+        assert_eq!(config.base_dir, base_dir);
+    }
+
+    #[tokio::test]
+    async fn test_knowledge_bases_dir_structure() {
+        let temp_dir = TempDir::new().unwrap();
+        let os = create_test_os(&temp_dir).await;
+
+        let base_dir = crate::util::directories::knowledge_bases_dir(&os).unwrap();
+
+        // Verify directory structure
+        assert!(base_dir.to_string_lossy().contains("knowledge_bases"));
+    }
+}
diff --git a/crates/semantic-search-client/Cargo.toml b/crates/semantic-search-client/Cargo.toml
index 9f6b09551..5024a5f4c 100644
--- a/crates/semantic-search-client/Cargo.toml
+++ b/crates/semantic-search-client/Cargo.toml
@@ -24,6 +24,7 @@ rayon.workspace = true
 tempfile.workspace = true
 tokio.workspace = true
 tokio-util.workspace = true
+glob.workspace = true
 
 # Vector search library - pin to avoid edition2024 requirement
 hnsw_rs = "=0.3.1"
diff --git a/crates/semantic-search-client/README.md b/crates/semantic-search-client/README.md
index 3a8e87ba8..c4d94f9ac 100644
--- a/crates/semantic-search-client/README.md
+++ b/crates/semantic-search-client/README.md
@@ -7,17 +7,20 @@ Rust library for managing semantic memory contexts with vector embeddings, enabl
 
 ## Features
 
+- **Async-First Design**: Built for modern async Rust applications with tokio
 - **Semantic Memory Management**: Create, store, and search through semantic memory contexts
+- **Pattern Filtering**: Include/exclude files using glob-style patterns during indexing
 - **Vector Embeddings**: Generate high-quality text embeddings for semantic similarity search
 - **Multi-Platform Support**: Works on macOS, Windows, and Linux with optimized backends
 - **Hardware Acceleration**: Uses Metal on macOS and optimized backends on other platforms
 - **File Processing**: Process various file types including text, markdown, JSON, and code
 - **Persistent Storage**: Save contexts to disk for long-term storage and retrieval
-- **Progress Tracking**: Detailed progress reporting for long-running operations
+- **Background Processing**: Non-blocking indexing with progress tracking and cancellation
 - **Parallel Processing**: Efficiently process large directories with parallel execution
 - **Memory Efficient**: Stream large files and directories without excessive memory usage
 - **Cross-Platform Compatibility**: Fallback mechanisms for all platforms and architectures
 - **🆕 Configurable File Limits**: Built-in protection against indexing too many files (default: 5,000 files)
+- **🆕 Database Settings Integration**: Configurable chunk sizes, overlap, and limits
 
 ## Installation
 
@@ -31,30 +34,38 @@ semantic_search_client = "0.1.0"
 ## Quick Start
 
 ```rust
-use semantic_search_client::{SemanticSearchClient, SemanticSearchConfig, Result};
-use std::path::Path;
+use semantic_search_client::{AsyncSemanticSearchClient, AddContextRequest, Result};
+use std::path::PathBuf;
 
-fn main() -> Result<()> {
-    // Create a new client with default settings (5,000 file limit)
-    let mut client = SemanticSearchClient::new_with_default_dir()?;
+#[tokio::main]
+async fn main() -> Result<()> {
+    // Create a new async client with default settings
+    let client = AsyncSemanticSearchClient::new_with_default_dir().await?;
     
-    // Add a context from a directory
-    let context_id = client.add_context_from_path(
-        Path::new("/path/to/project"),
-        "My Project",
-        "Code and documentation for my project",
-        true, // make it persistent
-        None, // no progress callback
-    )?;
+    // Add a context using the new structured request API
+    let request = AddContextRequest {
+        path: PathBuf::from("/path/to/project"),
+        name: "My Project".to_string(),
+        description: "Code and documentation for my project".to_string(),
+        persistent: true,
+        include_patterns: Some(vec!["**/*.rs".to_string(), "**/*.md".to_string()]),
+        exclude_patterns: Some(vec!["target/**".to_string(), "**/.git/**".to_string()]),
+    };
     
-    // Search within the context
-    let results = client.search_context(&context_id, "implement authentication", 5)?;
+    let (operation_id, _cancel_token) = client.add_context(request).await?;
+    println!("Started indexing with operation ID: {}", operation_id);
+    
+    // Search across all contexts
+    let results = client.search_all("implement authentication", None).await?;
     
     // Print the results
-    for result in results {
-        println!("Score: {}", result.distance);
-        if let Some(text) = result.text() {
-            println!("Text: {}", text);
+    for (context_id, context_results) in results {
+        println!("Results from context {}", context_id);
+        for result in context_results {
+            println!("Score: {}", result.distance);
+            if let Some(text) = result.text() {
+                println!("Text: {}", text);
+            }
         }
     }
     
@@ -126,29 +137,18 @@ The default selection logic prioritizes performance where possible:
 ### Creating a Client
 
 ```rust
-// With default settings (5,000 file limit)
-let client = SemanticSearchClient::new_with_default_dir()?;
+// Create async client with default settings
+let client = AsyncSemanticSearchClient::new_with_default_dir().await?;
 
 // With custom directory
-let client = SemanticSearchClient::new("/path/to/storage")?;
+let client = AsyncSemanticSearchClient::new("/path/to/storage").await?;
 
 // With custom configuration
 let config = SemanticSearchConfig::default()
     .set_max_files(10000)      // Allow up to 10,000 files
     .set_chunk_size(1024);     // Custom chunk size
-let client = SemanticSearchClient::with_config("/path/to/storage", config)?;
-
-// With specific embedding type
-use semantic_search_client::embedding::EmbeddingType;
-let client = SemanticSearchClient::new_with_embedding_type(EmbeddingType::Candle)?;
-
-// With both custom config and embedding type
-let config = SemanticSearchConfig::with_max_files(15000); // 15,000 file limit
-let client = SemanticSearchClient::with_config_and_embedding_type(
-    "/path/to/storage",
-    config,
-    EmbeddingType::Candle
-)?;
+let client = AsyncSemanticSearchClient::with_config("/path/to/storage", config).await?;
+```
 ```
 
 ### Configuration Options
@@ -210,52 +210,53 @@ let client = SemanticSearchClient::with_config(path, config)?;
 ### Adding Contexts
 
 ```rust
-// From a file
-let file_context_id = client.add_context_from_file(
-    "/path/to/document.md",
-    "Documentation",
-    "Project documentation",
-    true, // persistent
-    None, // no progress callback
-)?;
+use semantic_search_client::AddContextRequest;
+use std::path::PathBuf;
+
+// Add context with pattern filtering
+let request = AddContextRequest {
+    path: PathBuf::from("/path/to/codebase"),
+    name: "Rust Codebase".to_string(),
+    description: "Main Rust project source code".to_string(),
+    persistent: true,
+    include_patterns: Some(vec![
+        "**/*.rs".to_string(),
+        "**/*.toml".to_string(),
+        "**/*.md".to_string(),
+    ]),
+    exclude_patterns: Some(vec![
+        "target/**".to_string(),
+        "**/.git/**".to_string(),
+        "**/node_modules/**".to_string(),
+    ]),
+};
 
-// From a directory with progress reporting
-let dir_context_id = client.add_context_from_directory(
-    "/path/to/codebase",
-    "Codebase",
-    "Project source code",
-    true, // persistent
-    Some(|status| {
-        match status {
-            ProgressStatus::CountingFiles => println!("Counting files..."),
-            ProgressStatus::StartingIndexing(count) => println!("Starting indexing {} files", count),
-            ProgressStatus::Indexing(current, total) => 
-                println!("Indexing file {}/{}", current, total),
-            ProgressStatus::CreatingSemanticContext => 
-                println!("Creating semantic context..."),
-            ProgressStatus::GeneratingEmbeddings(current, total) => 
-                println!("Generating embeddings {}/{}", current, total),
-            ProgressStatus::BuildingIndex => println!("Building index..."),
-            ProgressStatus::Finalizing => println!("Finalizing..."),
-            ProgressStatus::Complete => println!("Indexing complete!"),
-        }
-    }),
-)?;
+let (operation_id, cancel_token) = client.add_context(request).await?;
+println!("Started indexing with operation ID: {}", operation_id);
+
+// Add context without pattern filtering
+let simple_request = AddContextRequest {
+    path: PathBuf::from("/path/to/docs"),
+    name: "Documentation".to_string(),
+    description: "Project documentation".to_string(),
+    persistent: true,
+    include_patterns: None,
+    exclude_patterns: None,
+};
 
-// From raw text
-let text_context_id = client.add_context_from_text(
-    "This is some text to remember",
-    "Note",
-    "Important information",
-    false, // volatile
-)?;
+let (operation_id, _) = client.add_context(simple_request).await?;
+
+// Monitor progress (the client runs indexing in background)
+let status = client.get_status_data().await?;
+println!("Active operations: {}", status.active_count);
+```
 ```
 
 ### Searching
 
 ```rust
 // Search across all contexts
-let all_results = client.search_all("authentication implementation", 5)?;
+let all_results = client.search_all("authentication implementation", None).await?;
 for (context_id, results) in all_results {
     println!("Results from context {}", context_id);
     for result in results {
@@ -266,39 +267,102 @@ for (context_id, results) in all_results {
     }
 }
 
-// Search in a specific context
-let context_results = client.search_context(
-    &context_id,
-    "authentication implementation",
-    5,
-)?;
+// Get all contexts first, then search specific ones
+let contexts = client.get_contexts().await;
+for context in &contexts {
+    println!("Available context: {} ({})", context.name, context.id);
+}
+```
 ```
 
 ### Managing Contexts
 
 ```rust
 // Get all contexts
-let contexts = client.get_all_contexts();
+let contexts = client.get_contexts().await;
 for context in contexts {
     println!("Context: {} ({})", context.name, context.id);
     println!("  Description: {}", context.description);
     println!("  Created: {}", context.created_at);
     println!("  Items: {}", context.item_count);
+    println!("  Include patterns: {:?}", context.include_patterns);
+    println!("  Exclude patterns: {:?}", context.exclude_patterns);
 }
 
-// Make a volatile context persistent
-client.make_persistent(
-    &context_id,
-    "Saved Context",
-    "Important information saved for later",
-)?;
+// Remove contexts
+client.remove_context_by_id("context-id").await?;
+client.remove_context_by_name("Context Name").await?;
+client.remove_context_by_path("/path/to/indexed/directory").await?;
+
+// Cancel ongoing operations
+client.cancel_operation(operation_id).await?;
+client.cancel_all_operations().await?;
+
+// Get system status
+let status = client.get_status_data().await?;
+println!("Total contexts: {}", status.total_contexts);
+println!("Active operations: {}", status.active_count);
+```
+```
 
-// Remove a context
-client.remove_context_by_id(&context_id, true)?; // true to delete persistent storage
-client.remove_context_by_name("My Context", true)?;
-client.remove_context_by_path("/path/to/indexed/directory", true)?;
+## Pattern Filtering
+
+The library supports glob-style pattern filtering to control which files are indexed:
+
+### Include Patterns
+Only files matching these patterns will be indexed:
+```rust
+let request = AddContextRequest {
+    // ... other fields
+    include_patterns: Some(vec![
+        "**/*.rs".to_string(),      // All Rust files
+        "**/*.md".to_string(),      // All Markdown files
+        "src/**/*.toml".to_string(), // TOML files in src directory
+    ]),
+    exclude_patterns: None,
+};
 ```
 
+### Exclude Patterns
+Files matching these patterns will be skipped:
+```rust
+let request = AddContextRequest {
+    // ... other fields
+    include_patterns: None,
+    exclude_patterns: Some(vec![
+        "target/**".to_string(),        // Build artifacts
+        "**/.git/**".to_string(),       // Git metadata
+        "**/node_modules/**".to_string(), // Node.js dependencies
+        "**/*.log".to_string(),         // Log files
+    ]),
+};
+```
+
+### Combined Filtering
+Use both include and exclude patterns for precise control:
+```rust
+let request = AddContextRequest {
+    // ... other fields
+    include_patterns: Some(vec![
+        "**/*.rs".to_string(),
+        "**/*.toml".to_string(),
+        "**/*.md".to_string(),
+    ]),
+    exclude_patterns: Some(vec![
+        "target/**".to_string(),
+        "**/tests/**".to_string(),  // Skip test files
+        "**/*_test.rs".to_string(), // Skip test files
+    ]),
+};
+```
+
+### Pattern Syntax
+- `**` matches any number of directories
+- `*` matches any characters within a single path segment
+- `?` matches a single character
+- `[abc]` matches any character in the set
+- `{a,b,c}` matches any of the alternatives
+
 ## Advanced Features
 
 ### Custom Embedding Models
@@ -327,6 +391,7 @@ let client = SemanticSearchClient::with_embedding_type(
     EmbeddingType::BM25,
 )?;
 ```
+```
 
 ### Parallel Processing
 
diff --git a/crates/semantic-search-client/src/client/async_implementation.rs b/crates/semantic-search-client/src/client/async_implementation.rs
index a5de69c1e..2efdf6a69 100644
--- a/crates/semantic-search-client/src/client/async_implementation.rs
+++ b/crates/semantic-search-client/src/client/async_implementation.rs
@@ -32,9 +32,11 @@ use crate::error::{
     SemanticSearchError,
 };
 use crate::types::{
+    AddContextRequest,
     ContextId,
     DataPoint,
     IndexingJob,
+    IndexingParams,
     KnowledgeContext,
     OperationHandle,
     OperationStatus,
@@ -84,9 +86,68 @@ struct BackgroundWorker {
 const MAX_CONCURRENT_OPERATIONS: usize = 3;
 
 impl AsyncSemanticSearchClient {
-    /// Create a new async semantic search client
+    /// Create a new async semantic search client with custom config
+    pub async fn with_config(base_dir: impl AsRef<Path>, config: SemanticSearchConfig) -> Result<Self> {
+        let base_dir = base_dir.as_ref().to_path_buf();
+
+        tokio::fs::create_dir_all(&base_dir).await?;
+
+        // Create models directory
+        crate::config::ensure_models_dir(&base_dir)?;
+
+        // Pre-download models if the embedding type requires them
+        Self::ensure_models_downloaded(&config.embedding_type).await?;
+
+        let embedder = embedder_factory::create_embedder(config.embedding_type)?;
+
+        // Load metadata for persistent contexts
+        let contexts_file = base_dir.join("contexts.json");
+        let persistent_contexts: HashMap<ContextId, KnowledgeContext> = utils::load_json_from_file(&contexts_file)?;
+
+        let contexts = Arc::new(RwLock::new(persistent_contexts));
+        let volatile_contexts = Arc::new(RwLock::new(HashMap::new()));
+        let active_operations = Arc::new(RwLock::new(HashMap::new()));
+        let (job_tx, job_rx) = mpsc::unbounded_channel();
+
+        // Start background worker
+        let worker_embedder = embedder_factory::create_embedder(config.embedding_type)?;
+        let worker = BackgroundWorker {
+            job_rx,
+            contexts: contexts.clone(),
+            volatile_contexts: volatile_contexts.clone(),
+            active_operations: active_operations.clone(),
+            embedder: worker_embedder,
+            config: config.clone(),
+            base_dir: base_dir.clone(),
+            indexing_semaphore: Arc::new(Semaphore::new(MAX_CONCURRENT_OPERATIONS)),
+        };
+
+        tokio::spawn(worker.run());
+
+        let mut client = Self {
+            base_dir,
+            contexts,
+            volatile_contexts,
+            embedder,
+            config,
+            job_tx,
+            active_operations,
+        };
+
+        // Load all persistent contexts
+        client.load_persistent_contexts().await?;
+
+        Ok(client)
+    }
+
+    /// Create a new async semantic search client with default config
     pub async fn new(base_dir: impl AsRef<Path>) -> Result<Self> {
-        Self::with_embedding_type(base_dir, EmbeddingType::default()).await
+        let base_dir_path = base_dir.as_ref().to_path_buf();
+        let config = SemanticSearchConfig {
+            base_dir: base_dir_path,
+            ..Default::default()
+        };
+        Self::with_config(base_dir, config).await
     }
 
     /// Create a new semantic search client with the default base directory
@@ -157,83 +218,28 @@ impl AsyncSemanticSearchClient {
         config::get_default_base_dir()
     }
 
-    /// Create a new async semantic search client with custom configuration and embedding type
-    pub async fn with_embedding_type(base_dir: impl AsRef<Path>, embedding_type: EmbeddingType) -> Result<Self> {
-        let base_dir = base_dir.as_ref().to_path_buf();
-        tokio::fs::create_dir_all(&base_dir).await?;
-
-        // Create models directory
-        config::ensure_models_dir(&base_dir)?;
-
-        // Initialize the configuration
-        if let Err(e) = config::init_config(&base_dir) {
-            tracing::error!("Failed to initialize semantic search configuration: {}", e);
-        }
-
-        // Pre-download models if the embedding type requires them
-        Self::ensure_models_downloaded(&embedding_type).await?;
-
-        let embedder = embedder_factory::create_embedder(embedding_type)?;
-
-        // Load metadata for persistent contexts
-        let contexts_file = base_dir.join("contexts.json");
-        let persistent_contexts: HashMap<ContextId, KnowledgeContext> = utils::load_json_from_file(&contexts_file)?;
-
-        let contexts = Arc::new(RwLock::new(persistent_contexts));
-        let volatile_contexts = Arc::new(RwLock::new(HashMap::new()));
-        let active_operations = Arc::new(RwLock::new(HashMap::new()));
-        let (job_tx, job_rx) = mpsc::unbounded_channel();
-
-        // Start background worker - we'll need to create a new embedder for the worker
-        // Models should already be downloaded by now
-        let worker_embedder = embedder_factory::create_embedder(embedding_type)?;
-        // Makes sure it respects configuration even if tweaked by user.
-        let loaded_config = config::get_config().clone();
-        let worker = BackgroundWorker {
-            job_rx,
-            contexts: contexts.clone(),
-            volatile_contexts: volatile_contexts.clone(),
-            active_operations: active_operations.clone(),
-            embedder: worker_embedder,
-            config: loaded_config.clone(),
-            base_dir: base_dir.clone(),
-            indexing_semaphore: Arc::new(Semaphore::new(MAX_CONCURRENT_OPERATIONS)),
-        };
-
-        tokio::spawn(worker.run());
-
-        let mut client = Self {
-            base_dir,
-            contexts,
-            volatile_contexts,
-            embedder,
-            config: loaded_config,
-            job_tx,
-            active_operations,
-        };
-
-        // Load all persistent contexts
-        client.load_persistent_contexts().await?;
-
-        Ok(client)
-    }
-
-    /// Add a context from a path (async, cancellable)
-    pub async fn add_context_from_path(
-        &self,
-        path: impl AsRef<Path>,
-        name: &str,
-        description: &str,
-        persistent: bool,
-    ) -> Result<(Uuid, CancellationToken)> {
-        let path = path.as_ref();
-        let canonical_path = path.canonicalize().map_err(|_e| {
-            SemanticSearchError::InvalidPath(format!("Path does not exist or is not accessible: {}", path.display()))
+    /// Add context using structured request
+    pub async fn add_context(&self, request: AddContextRequest) -> Result<(Uuid, CancellationToken)> {
+        let canonical_path = request.path.canonicalize().map_err(|_e| {
+            SemanticSearchError::InvalidPath(format!(
+                "Path does not exist or is not accessible: {}",
+                request.path.display()
+            ))
         })?;
 
         // Check for conflicts
         self.check_path_exists(&canonical_path).await?;
 
+        // Validate patterns early to fail fast
+        if let Some(ref include_patterns) = request.include_patterns {
+            crate::pattern_filter::PatternFilter::new(include_patterns, &[])
+                .map_err(|e| SemanticSearchError::InvalidArgument(format!("Invalid include pattern: {}", e)))?;
+        }
+        if let Some(ref exclude_patterns) = request.exclude_patterns {
+            crate::pattern_filter::PatternFilter::new(&[], exclude_patterns)
+                .map_err(|e| SemanticSearchError::InvalidArgument(format!("Invalid exclude pattern: {}", e)))?;
+        }
+
         let operation_id = Uuid::new_v4();
         let cancel_token = CancellationToken::new();
 
@@ -241,7 +247,7 @@ impl AsyncSemanticSearchClient {
         self.register_operation(
             operation_id,
             OperationType::Indexing {
-                name: name.to_string(),
+                name: request.name.clone(),
                 path: canonical_path.to_string_lossy().to_string(),
             },
             cancel_token.clone(),
@@ -253,9 +259,11 @@ impl AsyncSemanticSearchClient {
             id: operation_id,
             cancel: cancel_token.clone(),
             path: canonical_path,
-            name: name.to_string(),
-            description: description.to_string(),
-            persistent,
+            name: request.name,
+            description: request.description,
+            persistent: request.persistent,
+            include_patterns: request.include_patterns,
+            exclude_patterns: request.exclude_patterns,
         };
 
         self.job_tx
@@ -372,6 +380,33 @@ impl AsyncSemanticSearchClient {
         }
     }
 
+    /// Cancel the most recent operation
+    pub async fn cancel_most_recent_operation(&self) -> Result<String> {
+        let operations = self.active_operations.read().await;
+
+        if operations.is_empty() {
+            return Err(SemanticSearchError::OperationFailed(
+                "No active operations to cancel".to_string(),
+            ));
+        }
+
+        // Find the most recent operation (highest started_at time)
+        let most_recent = operations
+            .iter()
+            .max_by_key(|(_, handle)| handle.started_at)
+            .map(|(id, _)| *id);
+
+        drop(operations); // Release the read lock
+
+        if let Some(operation_id) = most_recent {
+            self.cancel_operation(operation_id).await
+        } else {
+            Err(SemanticSearchError::OperationFailed(
+                "No active operations found".to_string(),
+            ))
+        }
+    }
+
     /// Cancel all active operations
     pub async fn cancel_all_operations(&self) -> Result<String> {
         let mut operations = self.active_operations.write().await;
@@ -652,7 +687,8 @@ impl AsyncSemanticSearchClient {
         }
 
         // Create a new semantic context
-        let semantic_context = SemanticContext::new(context_dir.join("data.json"))?;
+        let data_file = context_dir.join("data.json");
+        let semantic_context = SemanticContext::new(data_file)?;
 
         // Store the semantic context
         let mut volatile_contexts = self.volatile_contexts.write().await;
@@ -770,6 +806,22 @@ impl AsyncSemanticSearchClient {
 
 // Background Worker Implementation
 impl BackgroundWorker {
+    /// Create pattern filter from include/exclude patterns
+    fn create_pattern_filter(
+        include_patterns: &Option<Vec<String>>,
+        exclude_patterns: &Option<Vec<String>>,
+    ) -> std::result::Result<Option<crate::pattern_filter::PatternFilter>, String> {
+        if include_patterns.is_some() || exclude_patterns.is_some() {
+            let inc = include_patterns.as_deref().unwrap_or(&[]);
+            let exc = exclude_patterns.as_deref().unwrap_or(&[]);
+            Ok(Some(
+                crate::pattern_filter::PatternFilter::new(inc, exc).map_err(|e| format!("Invalid patterns: {}", e))?,
+            ))
+        } else {
+            Ok(None)
+        }
+    }
+
     async fn run(mut self) {
         debug!("Background worker started for async semantic search client");
 
@@ -782,9 +834,18 @@ impl BackgroundWorker {
                     name,
                     description,
                     persistent,
+                    include_patterns,
+                    exclude_patterns,
                 } => {
-                    self.process_add_directory(id, path, name, description, persistent, cancel)
-                        .await;
+                    let params = IndexingParams {
+                        path,
+                        name,
+                        description,
+                        persistent,
+                        include_patterns,
+                        exclude_patterns,
+                    };
+                    self.process_add_directory(id, params, cancel).await;
                 },
                 IndexingJob::Clear { id, cancel } => {
                     self.process_clear(id, cancel).await;
@@ -795,16 +856,12 @@ impl BackgroundWorker {
         debug!("Background worker stopped");
     }
 
-    async fn process_add_directory(
-        &self,
-        operation_id: Uuid,
-        path: PathBuf,
-        name: String,
-        description: String,
-        persistent: bool,
-        cancel_token: CancellationToken,
-    ) {
-        debug!("Processing AddDirectory job: {} -> {}", name, path.display());
+    async fn process_add_directory(&self, operation_id: Uuid, params: IndexingParams, cancel_token: CancellationToken) {
+        debug!(
+            "Processing AddDirectory job: {} -> {}",
+            params.name,
+            params.path.display()
+        );
 
         if cancel_token.is_cancelled() {
             self.mark_operation_cancelled(operation_id).await;
@@ -846,9 +903,15 @@ impl BackgroundWorker {
         };
 
         // Perform actual indexing
-        let result = self
-            .perform_indexing(operation_id, path, name, description, persistent, cancel_token)
-            .await;
+        let indexing_params = IndexingParams {
+            path: params.path,
+            name: params.name,
+            description: params.description,
+            persistent: params.persistent,
+            include_patterns: params.include_patterns,
+            exclude_patterns: params.exclude_patterns,
+        };
+        let result = self.perform_indexing(operation_id, indexing_params, cancel_token).await;
 
         match result {
             Ok(context_id) => {
@@ -865,14 +928,11 @@ impl BackgroundWorker {
     async fn perform_indexing(
         &self,
         operation_id: Uuid,
-        path: PathBuf,
-        name: String,
-        description: String,
-        persistent: bool,
+        params: IndexingParams,
         cancel_token: CancellationToken,
     ) -> std::result::Result<String, String> {
-        if !path.exists() {
-            return Err(format!("Path '{}' does not exist", path.display()));
+        if !params.path.exists() {
+            return Err(format!("Path '{}' does not exist", params.path.display()));
         }
 
         // Check for cancellation before starting
@@ -893,7 +953,7 @@ impl BackgroundWorker {
         let cancel_token_clone = cancel_token.clone();
 
         // Create context directory
-        let context_dir = if persistent {
+        let context_dir = if params.persistent {
             base_dir.join(&context_id)
         } else {
             std::env::temp_dir().join("semantic_search").join(&context_id)
@@ -909,7 +969,14 @@ impl BackgroundWorker {
         }
 
         // Count files and notify progress
-        let file_count = self.count_files_in_directory(&path, operation_id).await?;
+        let file_count = self
+            .count_files_in_directory(
+                &params.path,
+                operation_id,
+                &params.include_patterns,
+                &params.exclude_patterns,
+            )
+            .await?;
 
         // Check if file count exceeds the configured limit
         if file_count > config.max_files {
@@ -941,7 +1008,14 @@ impl BackgroundWorker {
 
         // Process files with cancellation checks
         let items = self
-            .process_directory_files(&path, file_count, operation_id, &cancel_token_clone)
+            .process_directory_files(
+                &params.path,
+                file_count,
+                operation_id,
+                &cancel_token_clone,
+                &params.include_patterns,
+                &params.exclude_patterns,
+            )
             .await?;
 
         // Check cancellation before creating semantic context
@@ -972,7 +1046,7 @@ impl BackgroundWorker {
         }
 
         // Save context if persistent
-        if persistent {
+        if params.persistent {
             semantic_context
                 .save()
                 .map_err(|e| format!("Failed to save context: {}", e))?;
@@ -981,10 +1055,12 @@ impl BackgroundWorker {
         // Store the context
         self.store_context(
             &context_id,
-            &name,
-            &description,
-            persistent,
-            Some(path.to_string_lossy().to_string()),
+            &params.name,
+            &params.description,
+            params.persistent,
+            Some(params.path.to_string_lossy().to_string()),
+            &params.include_patterns,
+            &params.exclude_patterns,
             semantic_context,
             file_count,
         )
@@ -1228,6 +1304,8 @@ impl BackgroundWorker {
         description: &str,
         persistent: bool,
         source_path: Option<String>,
+        include_patterns: &Option<Vec<String>>,
+        exclude_patterns: &Option<Vec<String>>,
         semantic_context: SemanticContext,
         item_count: usize,
     ) -> std::result::Result<(), String> {
@@ -1238,6 +1316,10 @@ impl BackgroundWorker {
             description,
             persistent,
             source_path,
+            (
+                include_patterns.as_deref().unwrap_or(&[]).to_vec(),
+                exclude_patterns.as_deref().unwrap_or(&[]).to_vec(),
+            ),
             item_count,
         );
 
@@ -1266,6 +1348,8 @@ impl BackgroundWorker {
         &self,
         dir_path: &Path,
         operation_id: Uuid,
+        include_patterns: &Option<Vec<String>>,
+        exclude_patterns: &Option<Vec<String>>,
     ) -> std::result::Result<usize, String> {
         self.update_operation_status(operation_id, "Counting files...".to_string())
             .await;
@@ -1274,6 +1358,9 @@ impl BackgroundWorker {
         let dir_path = dir_path.to_path_buf();
         let active_operations = self.active_operations.clone();
 
+        // Create pattern filter if patterns are provided
+        let pattern_filter = Self::create_pattern_filter(include_patterns, exclude_patterns)?;
+
         let count_result = tokio::task::spawn_blocking(move || {
             let mut count = 0;
             let mut checked = 0;
@@ -1290,6 +1377,12 @@ impl BackgroundWorker {
                         .and_then(|n| n.to_str())
                         .is_some_and(|s| s.starts_with('.'))
                 })
+                .filter(|e| {
+                    // Apply pattern filter if present
+                    pattern_filter
+                        .as_ref()
+                        .is_none_or(|filter| filter.should_include(e.path()))
+                })
             {
                 count += 1;
                 checked += 1;
@@ -1334,12 +1427,17 @@ impl BackgroundWorker {
         file_count: usize,
         operation_id: Uuid,
         cancel_token: &CancellationToken,
+        include_patterns: &Option<Vec<String>>,
+        exclude_patterns: &Option<Vec<String>>,
     ) -> std::result::Result<Vec<serde_json::Value>, String> {
-        use crate::processing::process_file;
+        use crate::processing::process_file_with_config;
 
         self.update_operation_status(operation_id, format!("Starting indexing ({} files)", file_count))
             .await;
 
+        // Create pattern filter if patterns are provided
+        let pattern_filter = Self::create_pattern_filter(include_patterns, exclude_patterns)?;
+
         let mut processed_files = 0;
         let mut items = Vec::new();
 
@@ -1348,6 +1446,12 @@ impl BackgroundWorker {
             .into_iter()
             .filter_map(|e| e.ok())
             .filter(|e| e.file_type().is_file())
+            .filter(|e| {
+                // Apply pattern filter if present
+                pattern_filter
+                    .as_ref()
+                    .is_none_or(|filter| filter.should_include(e.path()))
+            })
         {
             // Check for cancellation frequently
             if cancel_token.is_cancelled() {
@@ -1366,7 +1470,7 @@ impl BackgroundWorker {
             }
 
             // Process the file
-            match process_file(path) {
+            match process_file_with_config(path, Some(self.config.chunk_size), Some(self.config.chunk_overlap)) {
                 Ok(mut file_items) => items.append(&mut file_items),
                 Err(_) => continue, // Skip files that fail to process
             }
diff --git a/crates/semantic-search-client/src/client/implementation.rs b/crates/semantic-search-client/src/client/implementation.rs
index aec38c059..2120782c5 100644
--- a/crates/semantic-search-client/src/client/implementation.rs
+++ b/crates/semantic-search-client/src/client/implementation.rs
@@ -17,15 +17,12 @@ use crate::client::{
     utils,
 };
 use crate::config;
-use crate::embedding::{
-    EmbeddingType,
-    TextEmbedderTrait,
-};
+use crate::embedding::TextEmbedderTrait;
 use crate::error::{
     Result,
     SemanticSearchError,
 };
-use crate::processing::process_file;
+use crate::processing::process_file_with_config;
 use crate::types::{
     ContextId,
     ContextMap,
@@ -84,39 +81,25 @@ impl SemanticSearchClient {
     ///
     /// A new SemanticSearchClient instance
     pub fn new(base_dir: impl AsRef<Path>) -> Result<Self> {
-        Self::with_embedding_type(base_dir, EmbeddingType::default())
-    }
-
-    /// Create a new semantic search client with a specific embedding type
-    ///
-    /// # Arguments
-    ///
-    /// * `base_dir` - Base directory for storing persistent contexts
-    /// * `embedding_type` - Type of embedding engine to use
-    ///
-    /// # Returns
-    ///
-    /// A new SemanticSearchClient instance
-    pub fn with_embedding_type(base_dir: impl AsRef<Path>, embedding_type: EmbeddingType) -> Result<Self> {
-        Self::with_config_and_embedding_type(base_dir, crate::config::SemanticSearchConfig::default(), embedding_type)
+        let base_dir_path = base_dir.as_ref().to_path_buf();
+        let config = crate::config::SemanticSearchConfig {
+            base_dir: base_dir_path,
+            ..Default::default()
+        };
+        Self::with_config(base_dir, config)
     }
 
-    /// Create a new semantic search client with custom configuration and embedding type
+    /// Create a new semantic search client with custom configuration
     ///
     /// # Arguments
     ///
     /// * `base_dir` - Base directory for storing persistent contexts
     /// * `config` - Configuration for the client
-    /// * `embedding_type` - Type of embedding engine to use
     ///
     /// # Returns
     ///
     /// A new SemanticSearchClient instance
-    pub fn with_config_and_embedding_type(
-        base_dir: impl AsRef<Path>,
-        config: crate::config::SemanticSearchConfig,
-        embedding_type: EmbeddingType,
-    ) -> Result<Self> {
+    pub fn with_config(base_dir: impl AsRef<Path>, config: crate::config::SemanticSearchConfig) -> Result<Self> {
         let base_dir = base_dir.as_ref().to_path_buf();
         fs::create_dir_all(&base_dir)?;
 
@@ -129,7 +112,7 @@ impl SemanticSearchClient {
             // Continue with default config if initialization fails
         }
 
-        let embedder = embedder_factory::create_embedder(embedding_type)?;
+        let embedder = embedder_factory::create_embedder(config.embedding_type)?;
 
         // Load metadata for persistent contexts
         let contexts_file = base_dir.join("contexts.json");
@@ -155,20 +138,6 @@ impl SemanticSearchClient {
         Ok(client)
     }
 
-    /// Create a new semantic search client with custom configuration
-    ///
-    /// # Arguments
-    ///
-    /// * `base_dir` - Base directory for storing persistent contexts
-    /// * `config` - Configuration for the client
-    ///
-    /// # Returns
-    ///
-    /// A new SemanticSearchClient instance
-    pub fn with_config(base_dir: impl AsRef<Path>, config: crate::config::SemanticSearchConfig) -> Result<Self> {
-        Self::with_config_and_embedding_type(base_dir, config, EmbeddingType::default())
-    }
-
     /// Get the default base directory for memory bank
     ///
     /// # Returns
@@ -201,21 +170,6 @@ impl SemanticSearchClient {
         Self::new(base_dir)
     }
 
-    /// Create a new semantic search client with the default base directory and specific embedding
-    /// type
-    ///
-    /// # Arguments
-    ///
-    /// * `embedding_type` - Type of embedding engine to use
-    ///
-    /// # Returns
-    ///
-    /// A new SemanticSearchClient instance
-    pub fn new_with_embedding_type(embedding_type: EmbeddingType) -> Result<Self> {
-        let base_dir = Self::get_default_base_dir();
-        Self::with_embedding_type(base_dir, embedding_type)
-    }
-
     /// Get the current semantic search configuration
     ///
     /// # Returns
@@ -341,7 +295,7 @@ impl SemanticSearchClient {
         }
 
         // Process the file
-        let items = process_file(file_path)?;
+        let items = process_file_with_config(file_path, Some(self.config.chunk_size), Some(self.config.chunk_overlap))?;
 
         // Notify progress: Indexing
         if let Some(ref callback) = progress_callback {
@@ -418,7 +372,7 @@ impl SemanticSearchClient {
         }
 
         // Process files
-        let items = Self::process_directory_files(dir_path, file_count, &progress_callback)?;
+        let items = Self::process_directory_files(dir_path, file_count, &progress_callback, &self.config)?;
 
         // Create and populate semantic context
         let semantic_context = self.create_semantic_context(&context_dir, &items, &progress_callback)?;
@@ -454,6 +408,7 @@ impl SemanticSearchClient {
         dir_path: &Path,
         file_count: usize,
         progress_callback: &Option<F>,
+        config: &crate::config::SemanticSearchConfig,
     ) -> Result<Vec<Value>>
     where
         F: Fn(ProgressStatus) + Send + 'static,
@@ -485,7 +440,7 @@ impl SemanticSearchClient {
             }
 
             // Process the file
-            match process_file(path) {
+            match process_file_with_config(path, Some(config.chunk_size), Some(config.chunk_overlap)) {
                 Ok(mut file_items) => items.append(&mut file_items),
                 Err(_) => continue, // Skip files that fail to process
             }
@@ -576,7 +531,15 @@ impl SemanticSearchClient {
         }
 
         // Create the context metadata
-        let context = KnowledgeContext::new(id.to_string(), name, description, persistent, source_path, item_count);
+        let context = KnowledgeContext::new(
+            id.to_string(),
+            name,
+            description,
+            persistent,
+            source_path,
+            (vec![], vec![]),
+            item_count,
+        );
 
         // Store the context
         if persistent {
@@ -729,6 +692,7 @@ impl SemanticSearchClient {
                     "Temporary memory context",
                     false,
                     None,
+                    (vec![], vec![]),
                     0,
                 );
                 contexts.push(context);
@@ -908,6 +872,7 @@ impl SemanticSearchClient {
             context_description,
             true,
             None,
+            (vec![], vec![]),
             context_guard.get_data_points().len(),
         );
 
diff --git a/crates/semantic-search-client/src/config.rs b/crates/semantic-search-client/src/config.rs
index 03ce5b2e8..030566bbf 100644
--- a/crates/semantic-search-client/src/config.rs
+++ b/crates/semantic-search-client/src/config.rs
@@ -16,6 +16,8 @@ use serde::{
     Serialize,
 };
 
+use crate::embedding::EmbeddingType;
+
 /// Main configuration structure for the semantic search client.
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct SemanticSearchConfig {
@@ -42,6 +44,9 @@ pub struct SemanticSearchConfig {
 
     /// Base URL for hosted models
     pub hosted_models_base_url: String,
+
+    /// Embedding engine type to use
+    pub embedding_type: EmbeddingType,
 }
 
 impl SemanticSearchConfig {
@@ -87,6 +92,7 @@ impl Default for SemanticSearchConfig {
             base_dir: get_default_base_dir(),
             max_files: 10000, // Default limit of 10000 files
             hosted_models_base_url: "https://desktop-release.q.us-east-1.amazonaws.com/models".to_string(),
+            embedding_type: EmbeddingType::default(),
         }
     }
 }
@@ -187,13 +193,9 @@ pub fn init_config(base_dir: &Path) -> std::io::Result<()> {
 ///
 /// # Returns
 ///
-/// A reference to the global configuration
-///
-/// # Panics
-///
-/// Panics if the configuration has not been initialized
+/// A reference to the global configuration, or default if not initialized
 pub fn get_config() -> &'static SemanticSearchConfig {
-    CONFIG.get().expect("Semantic search configuration not initialized")
+    CONFIG.get_or_init(SemanticSearchConfig::default)
 }
 
 /// Loads the configuration from a file or creates a new one with default values.
@@ -338,6 +340,7 @@ mod tests {
             base_dir: temp_dir.path().to_path_buf(),
             max_files: 10000,
             hosted_models_base_url: "http://test.example.com/models".to_string(),
+            embedding_type: EmbeddingType::default(),
         };
 
         // Update the config
diff --git a/crates/semantic-search-client/src/embedding/trait_def.rs b/crates/semantic-search-client/src/embedding/trait_def.rs
index 1be80b675..80e694381 100644
--- a/crates/semantic-search-client/src/embedding/trait_def.rs
+++ b/crates/semantic-search-client/src/embedding/trait_def.rs
@@ -1,7 +1,12 @@
+use serde::{
+    Deserialize,
+    Serialize,
+};
+
 use crate::error::Result;
 
 /// Embedding engine type to use
-#[derive(Debug, Clone, Copy)]
+#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
 pub enum EmbeddingType {
     /// Use Candle embedding engine (not available on Linux ARM)
     #[cfg(not(all(target_os = "linux", target_arch = "aarch64")))]
diff --git a/crates/semantic-search-client/src/lib.rs b/crates/semantic-search-client/src/lib.rs
index cc4a6f764..c99a147b9 100644
--- a/crates/semantic-search-client/src/lib.rs
+++ b/crates/semantic-search-client/src/lib.rs
@@ -16,6 +16,8 @@ pub mod error;
 pub mod index;
 /// Model validation for SHA verification
 pub mod model_validator;
+/// Pattern filtering for file selection
+pub mod pattern_filter;
 /// File processing utilities
 pub mod processing;
 /// Data types for semantic search operations
diff --git a/crates/semantic-search-client/src/pattern_filter.rs b/crates/semantic-search-client/src/pattern_filter.rs
new file mode 100644
index 000000000..1c6821af3
--- /dev/null
+++ b/crates/semantic-search-client/src/pattern_filter.rs
@@ -0,0 +1,309 @@
+use std::path::Path;
+
+use glob::Pattern;
+
+/// Pattern-based file filtering for semantic search indexing
+#[derive(Debug, Clone)]
+pub struct PatternFilter {
+    include_patterns: Vec<Pattern>,
+    exclude_patterns: Vec<Pattern>,
+}
+
+impl PatternFilter {
+    /// Create a new pattern filter
+    pub fn new(include_patterns: &[String], exclude_patterns: &[String]) -> Result<Self, String> {
+        let include_patterns = include_patterns
+            .iter()
+            .map(|p| Pattern::new(p).map_err(|e| format!("Invalid include pattern '{}': {}", p, e)))
+            .collect::<Result<Vec<_>, _>>()?;
+
+        let exclude_patterns = exclude_patterns
+            .iter()
+            .map(|p| Pattern::new(p).map_err(|e| format!("Invalid exclude pattern '{}': {}", p, e)))
+            .collect::<Result<Vec<_>, _>>()?;
+
+        Ok(Self {
+            include_patterns,
+            exclude_patterns,
+        })
+    }
+
+    /// Check if a file should be included based on patterns
+    /// Handles both absolute and relative paths automatically
+    pub fn should_include(&self, file_path: &Path) -> bool {
+        // Check include patterns (if any)
+        if !self.include_patterns.is_empty() {
+            let matches_include = self
+                .include_patterns
+                .iter()
+                .any(|pattern| Self::matches_pattern(pattern, file_path));
+            if !matches_include {
+                return false;
+            }
+        }
+
+        // Check exclude patterns (if any)
+        if !self.exclude_patterns.is_empty() {
+            let matches_exclude = self
+                .exclude_patterns
+                .iter()
+                .any(|pattern| Self::matches_pattern(pattern, file_path));
+            if matches_exclude {
+                return false;
+            }
+        }
+
+        true
+    }
+
+    /// Match a pattern against a path, handling both absolute and relative paths
+    fn matches_pattern(pattern: &Pattern, file_path: &Path) -> bool {
+        let path_str = file_path.to_string_lossy();
+
+        // Try direct match first (for relative paths)
+        if pattern.matches(&path_str) {
+            return true;
+        }
+
+        // For absolute paths, try matching against path components
+        // This handles cases where pattern is "node_modules/**" but path is
+        // "/full/path/to/node_modules/file"
+        let components: Vec<_> = file_path
+            .components()
+            .map(|c| c.as_os_str().to_string_lossy())
+            .collect();
+
+        // Try to find a suffix of the path that matches the pattern
+        for i in 0..components.len() {
+            let suffix_path = components[i..].join("/");
+            if pattern.matches(&suffix_path) {
+                return true;
+            }
+        }
+
+        false
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::path::PathBuf;
+
+    use super::*;
+
+    #[test]
+    fn test_pattern_filter_creation() {
+        let filter = PatternFilter::new(&["*.rs".to_string()], &["target/**".to_string()]);
+        assert!(filter.is_ok());
+
+        let invalid_filter = PatternFilter::new(&["[".to_string()], &[]);
+        assert!(invalid_filter.is_err());
+        assert!(invalid_filter.unwrap_err().contains("Invalid include pattern"));
+    }
+
+    #[test]
+    fn test_include_patterns_work() {
+        // Test that include patterns work correctly
+        let include_patterns = vec!["*.rs".to_string()];
+        let exclude_patterns = vec![];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Should include .rs files
+        assert!(filter.should_include(&PathBuf::from("main.rs")));
+        assert!(filter.should_include(&PathBuf::from("lib.rs")));
+
+        // Should not include other files
+        assert!(!filter.should_include(&PathBuf::from("main.py")));
+        assert!(!filter.should_include(&PathBuf::from("README.md")));
+    }
+
+    #[test]
+    fn test_exclude_patterns_work() {
+        // Test that exclude patterns work correctly
+        let include_patterns = vec![];
+        let exclude_patterns = vec!["node_modules/**".to_string()];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Should exclude files in node_modules
+        assert!(!filter.should_include(&PathBuf::from("node_modules/package/index.js")));
+        assert!(!filter.should_include(&PathBuf::from("node_modules/lib.rs")));
+
+        // Should include other files
+        assert!(filter.should_include(&PathBuf::from("src/main.rs")));
+        assert!(filter.should_include(&PathBuf::from("README.md")));
+    }
+
+    #[test]
+    fn test_recursive_patterns() {
+        // Test that recursive patterns (**) work correctly
+        let include_patterns = vec!["**/*.rs".to_string()];
+        let exclude_patterns = vec![];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Should include .rs files at any depth
+        assert!(filter.should_include(&PathBuf::from("main.rs")));
+        assert!(filter.should_include(&PathBuf::from("src/main.rs")));
+        assert!(filter.should_include(&PathBuf::from("src/lib/mod.rs")));
+        assert!(filter.should_include(&PathBuf::from("deep/nested/path/file.rs")));
+
+        // Should not include non-.rs files
+        assert!(!filter.should_include(&PathBuf::from("src/main.py")));
+        assert!(!filter.should_include(&PathBuf::from("deep/nested/README.md")));
+    }
+
+    #[test]
+    fn test_combined_include_exclude() {
+        // Test that include and exclude patterns work together
+        let include_patterns = vec!["**/*.rs".to_string()];
+        let exclude_patterns = vec!["target/**".to_string()];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Should include .rs files not in target
+        assert!(filter.should_include(&PathBuf::from("src/main.rs")));
+        assert!(filter.should_include(&PathBuf::from("tests/test.rs")));
+
+        // Should exclude .rs files in target
+        assert!(!filter.should_include(&PathBuf::from("target/debug/main.rs")));
+        assert!(!filter.should_include(&PathBuf::from("target/release/lib.rs")));
+
+        // Should exclude non-.rs files everywhere
+        assert!(!filter.should_include(&PathBuf::from("src/main.py")));
+        assert!(!filter.should_include(&PathBuf::from("README.md")));
+    }
+
+    #[test]
+    fn test_node_modules_exclusion_issue() {
+        // Test the specific issue mentioned in PR: node_modules exclusion not working
+        let include_patterns = vec![];
+        let exclude_patterns = vec!["node_modules/**".to_string()];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // These should be excluded (the reported bug)
+        assert!(!filter.should_include(&PathBuf::from("node_modules/package.json")));
+        assert!(!filter.should_include(&PathBuf::from("node_modules/lib/index.js")));
+        assert!(!filter.should_include(&PathBuf::from("node_modules/deep/nested/file.txt")));
+
+        // These should be included
+        assert!(filter.should_include(&PathBuf::from("src/index.js")));
+        assert!(filter.should_include(&PathBuf::from("package.json")));
+    }
+
+    #[test]
+    fn test_node_modules_exclusion_with_temp_dir() {
+        use std::fs;
+
+        use tempfile::TempDir;
+
+        let temp_dir = TempDir::new().unwrap();
+        let temp_path = temp_dir.path();
+
+        // Create the directory structure
+        fs::create_dir_all(temp_path.join("node_modules/some-package")).unwrap();
+        fs::create_dir_all(temp_path.join("src")).unwrap();
+
+        // Create files
+        fs::write(temp_path.join("node_modules/package.json"), "{}").unwrap();
+        fs::write(temp_path.join("node_modules/some-package/index.js"), "// code").unwrap();
+        fs::write(temp_path.join("src/main.js"), "// main code").unwrap();
+        fs::write(temp_path.join("package.json"), "{}").unwrap();
+
+        // Test the filter
+        let include_patterns = vec![];
+        let exclude_patterns = vec!["node_modules/**".to_string()];
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Test relative to temp directory
+        let node_modules_file = PathBuf::from("node_modules/package.json");
+        let node_modules_nested = PathBuf::from("node_modules/some-package/index.js");
+        let src_file = PathBuf::from("src/main.js");
+        let root_file = PathBuf::from("package.json");
+
+        // These should be excluded (the reported bug)
+        assert!(
+            !filter.should_include(&node_modules_file),
+            "node_modules/package.json should be excluded"
+        );
+        assert!(
+            !filter.should_include(&node_modules_nested),
+            "node_modules/some-package/index.js should be excluded"
+        );
+
+        // These should be included
+        assert!(filter.should_include(&src_file), "src/main.js should be included");
+        assert!(filter.should_include(&root_file), "package.json should be included");
+    }
+
+    #[test]
+    fn test_pattern_documentation_accuracy() {
+        let filter = PatternFilter::new(&["*.rs".to_string()], &[]).unwrap();
+
+        assert!(filter.should_include(&PathBuf::from("main.rs"))); // Current dir - should match
+
+        // These should NOT match with *.rs (only * not **)
+        // If they do match, then the documentation is wrong
+        let _nested_matches = filter.should_include(&PathBuf::from("src/main.rs"));
+        assert!(_nested_matches, "*.rs should match nested files recursively");
+    }
+
+    #[test]
+    fn test_empty_patterns() {
+        // Test behavior with no patterns (should include everything)
+        let include_patterns = vec![];
+        let exclude_patterns = vec![];
+
+        let filter = PatternFilter::new(&include_patterns, &exclude_patterns).unwrap();
+
+        // Should include everything when no patterns are specified
+        assert!(filter.should_include(&PathBuf::from("main.rs")));
+        assert!(filter.should_include(&PathBuf::from("src/main.rs")));
+        assert!(filter.should_include(&PathBuf::from("node_modules/package.json")));
+        assert!(filter.should_include(&PathBuf::from("target/debug/main")));
+    }
+
+    #[test]
+    fn test_absolute_vs_relative_path_handling() {
+        use std::fs;
+
+        use tempfile::TempDir;
+
+        let temp_dir = TempDir::new().unwrap();
+        let temp_path = temp_dir.path();
+
+        // Create directory structure
+        fs::create_dir_all(temp_path.join("node_modules")).unwrap();
+        fs::create_dir_all(temp_path.join("src")).unwrap();
+
+        let filter = PatternFilter::new(&[], &["node_modules/**".to_string()]).unwrap();
+
+        // Test relative paths (should work)
+        let relative_excluded = PathBuf::from("node_modules/package.json");
+        let relative_included = PathBuf::from("src/main.js");
+
+        assert!(
+            !filter.should_include(&relative_excluded),
+            "Relative node_modules path should be excluded"
+        );
+        assert!(
+            filter.should_include(&relative_included),
+            "Relative src path should be included"
+        );
+
+        // Test absolute paths (the fix - should also work now)
+        let absolute_excluded = temp_path.join("node_modules/package.json");
+        let absolute_included = temp_path.join("src/main.js");
+
+        assert!(
+            !filter.should_include(&absolute_excluded),
+            "Absolute node_modules path should be excluded"
+        );
+        assert!(
+            filter.should_include(&absolute_included),
+            "Absolute src path should be included"
+        );
+    }
+}
diff --git a/crates/semantic-search-client/src/processing/file_processor.rs b/crates/semantic-search-client/src/processing/file_processor.rs
index df4d23fa5..c56e50d88 100644
--- a/crates/semantic-search-client/src/processing/file_processor.rs
+++ b/crates/semantic-search-client/src/processing/file_processor.rs
@@ -12,10 +12,36 @@ use crate::types::FileType;
 
 /// Determine the file type based on extension
 pub fn get_file_type(path: &Path) -> FileType {
-    match path.extension().and_then(|ext| ext.to_str()) {
+    match path
+        .extension()
+        .and_then(|ext| ext.to_str())
+        .map(|s| s.to_lowercase())
+        .as_deref()
+    {
+        // Plain text files
         Some("txt") => FileType::Text,
-        Some("md" | "markdown") => FileType::Markdown,
-        Some("json") => FileType::Json,
+
+        // Markdown files (including MDX)
+        Some("md" | "markdown" | "mdx") => FileType::Markdown,
+
+        // JSON files - now treated as text for better searchability
+        Some("json") => FileType::Text,
+
+        // Configuration files
+        Some("ini" | "conf" | "cfg" | "properties" | "env") => FileType::Text,
+
+        // Data files
+        Some("csv" | "tsv") => FileType::Text,
+
+        // Log files
+        Some("log") => FileType::Text,
+
+        // Documentation formats
+        Some("rtf" | "tex" | "rst") => FileType::Text,
+
+        // Web and markup formats (text-based)
+        Some("svg") => FileType::Text,
+
         // Code file extensions
         Some("rs") => FileType::Code,
         Some("py") => FileType::Code,
@@ -34,12 +60,23 @@ pub fn get_file_type(path: &Path) -> FileType {
         Some("sql") => FileType::Code,
         Some("yaml" | "yml") => FileType::Code,
         Some("toml") => FileType::Code,
-        // Default to unknown
+
+        // Handle files without extensions (common project files)
+        None => match path.file_name().and_then(|name| name.to_str()) {
+            Some("Dockerfile" | "Makefile" | "LICENSE" | "CHANGELOG" | "README") => FileType::Text,
+            Some(name) if name.starts_with('.') => match name {
+                ".gitignore" | ".env" | ".dockerignore" => FileType::Text,
+                _ => FileType::Unknown,
+            },
+            _ => FileType::Unknown,
+        },
+
+        // Default to unknown (includes office docs, PDFs, etc.)
         _ => FileType::Unknown,
     }
 }
 
-/// Process a file and extract its content
+/// Process a file and extract its content (backward compatible version)
 ///
 /// # Arguments
 ///
@@ -49,6 +86,25 @@ pub fn get_file_type(path: &Path) -> FileType {
 ///
 /// A vector of JSON objects representing the file content
 pub fn process_file(path: &Path) -> Result<Vec<Value>> {
+    process_file_with_config(path, None, None)
+}
+
+/// Process a file with custom chunk configuration
+///
+/// # Arguments
+///
+/// * `path` - Path to the file
+/// * `chunk_size` - Optional chunk size (uses default if None)
+/// * `chunk_overlap` - Optional chunk overlap (uses default if None)
+///
+/// # Returns
+///
+/// A vector of JSON objects representing the file content
+pub fn process_file_with_config(
+    path: &Path,
+    chunk_size: Option<usize>,
+    chunk_overlap: Option<usize>,
+) -> Result<Vec<Value>> {
     if !path.exists() {
         return Err(SemanticSearchError::InvalidPath(format!(
             "File does not exist: {}",
@@ -65,10 +121,10 @@ pub fn process_file(path: &Path) -> Result<Vec<Value>> {
     })?;
 
     match file_type {
-        FileType::Text | FileType::Markdown | FileType::Code => {
-            // For text-based files, chunk the content and create multiple data points
+        FileType::Text | FileType::Markdown | FileType::Code | FileType::Json => {
+            // For text-based files (including JSON), chunk the content and create multiple data points
             // Use the configured chunk size and overlap
-            let chunks = chunk_text(&content, None, None);
+            let chunks = chunk_text(&content, chunk_size, chunk_overlap);
             let path_str = path.to_string_lossy().to_string();
             let file_type_str = format!("{:?}", file_type);
 
@@ -112,22 +168,6 @@ pub fn process_file(path: &Path) -> Result<Vec<Value>> {
 
             Ok(results)
         },
-        FileType::Json => {
-            // For JSON files, parse the content
-            let json: Value =
-                serde_json::from_str(&content).map_err(|e| SemanticSearchError::SerializationError(e.to_string()))?;
-
-            match json {
-                Value::Array(items) => {
-                    // If it's an array, return each item
-                    Ok(items)
-                },
-                _ => {
-                    // Otherwise, return the whole object
-                    Ok(vec![json])
-                },
-            }
-        },
         FileType::Unknown => {
             // For unknown file types, just store the path
             let mut metadata = serde_json::Map::new();
@@ -144,11 +184,17 @@ pub fn process_file(path: &Path) -> Result<Vec<Value>> {
 /// # Arguments
 ///
 /// * `dir_path` - Path to the directory
+/// * `chunk_size` - Optional chunk size (uses default if None)
+/// * `chunk_overlap` - Optional chunk overlap (uses default if None)
 ///
 /// # Returns
 ///
 /// A vector of JSON objects representing the content of all files
-pub fn process_directory(dir_path: &Path) -> Result<Vec<Value>> {
+pub fn process_directory(
+    dir_path: &Path,
+    chunk_size: Option<usize>,
+    chunk_overlap: Option<usize>,
+) -> Result<Vec<Value>> {
     let mut results = Vec::new();
 
     for entry in walkdir::WalkDir::new(dir_path)
@@ -169,10 +215,75 @@ pub fn process_directory(dir_path: &Path) -> Result<Vec<Value>> {
         }
 
         // Process the file
-        if let Ok(mut items) = process_file(path) {
+        if let Ok(mut items) = process_file_with_config(path, chunk_size, chunk_overlap) {
             results.append(&mut items);
         }
     }
 
     Ok(results)
 }
+
+#[cfg(test)]
+mod tests {
+    use std::path::PathBuf;
+
+    use super::*;
+    use crate::types::FileType;
+
+    #[test]
+    fn test_file_type_detection() {
+        let test_cases = [
+            // Code files
+            ("main.rs", FileType::Code),
+            ("script.py", FileType::Code),
+            ("app.js", FileType::Code),
+            ("component.tsx", FileType::Code),
+            ("Main.java", FileType::Code),
+            ("main.c", FileType::Code),
+            ("index.html", FileType::Code),
+            ("styles.css", FileType::Code),
+            ("config.yaml", FileType::Code),
+            ("Cargo.toml", FileType::Code),
+            // Markdown files
+            ("README.md", FileType::Markdown),
+            ("doc.markdown", FileType::Markdown),
+            ("component.mdx", FileType::Markdown),
+            // Text files
+            ("notes.txt", FileType::Text),
+            ("data.json", FileType::Text),
+            ("config.ini", FileType::Text),
+            ("data.csv", FileType::Text),
+            ("Dockerfile", FileType::Text),
+            ("LICENSE", FileType::Text),
+            (".gitignore", FileType::Text),
+            // Case insensitive
+            ("Main.RS", FileType::Code),
+            ("README.MD", FileType::Markdown),
+            ("notes.TXT", FileType::Text),
+            // Unknown files
+            ("image.png", FileType::Unknown),
+            ("document.pdf", FileType::Unknown),
+            ("binary.exe", FileType::Unknown),
+            ("unknown_file", FileType::Unknown),
+        ];
+
+        for (filename, expected) in test_cases {
+            assert_eq!(
+                get_file_type(&PathBuf::from(filename)),
+                expected,
+                "Failed for {}",
+                filename
+            );
+        }
+    }
+
+    #[test]
+    fn test_unknown_file_types() {
+        // Binary files and unsupported formats
+        assert_eq!(get_file_type(&PathBuf::from("image.png")), FileType::Unknown);
+        assert_eq!(get_file_type(&PathBuf::from("document.pdf")), FileType::Unknown);
+        assert_eq!(get_file_type(&PathBuf::from("archive.zip")), FileType::Unknown);
+        assert_eq!(get_file_type(&PathBuf::from("binary.exe")), FileType::Unknown);
+        assert_eq!(get_file_type(&PathBuf::from("data.db")), FileType::Unknown);
+    }
+}
diff --git a/crates/semantic-search-client/src/processing/mod.rs b/crates/semantic-search-client/src/processing/mod.rs
index 393f82700..87d8b5dc5 100644
--- a/crates/semantic-search-client/src/processing/mod.rs
+++ b/crates/semantic-search-client/src/processing/mod.rs
@@ -7,5 +7,6 @@ pub use file_processor::{
     get_file_type,
     process_directory,
     process_file,
+    process_file_with_config,
 };
 pub use text_chunker::chunk_text;
diff --git a/crates/semantic-search-client/src/processing/text_chunker.rs b/crates/semantic-search-client/src/processing/text_chunker.rs
index 369128a3c..4d28af567 100644
--- a/crates/semantic-search-client/src/processing/text_chunker.rs
+++ b/crates/semantic-search-client/src/processing/text_chunker.rs
@@ -61,6 +61,7 @@ mod tests {
                     base_dir: std::path::PathBuf::from("."),
                     max_files: 1000, // Add missing max_files field
                     hosted_models_base_url: "http://test.example.com/models".to_string(),
+                    embedding_type: crate::embedding::EmbeddingType::default(),
                 };
                 // Use a different approach that doesn't access private static
                 let _ = crate::config::init_config(&std::env::temp_dir());
diff --git a/crates/semantic-search-client/src/types.rs b/crates/semantic-search-client/src/types.rs
index 1789db216..4ebcf50c3 100644
--- a/crates/semantic-search-client/src/types.rs
+++ b/crates/semantic-search-client/src/types.rs
@@ -17,6 +17,40 @@ use serde::{
 use tokio_util::sync::CancellationToken;
 use uuid::Uuid;
 
+/// Request for adding a new context to the knowledge base
+#[derive(Debug, Clone)]
+pub struct AddContextRequest {
+    /// Path to the directory or file to index
+    pub path: PathBuf,
+    /// Human-readable name for the context
+    pub name: String,
+    /// Description of the context
+    pub description: String,
+    /// Whether this context should be persistent
+    pub persistent: bool,
+    /// Optional patterns to include during indexing
+    pub include_patterns: Option<Vec<String>>,
+    /// Optional patterns to exclude during indexing
+    pub exclude_patterns: Option<Vec<String>>,
+}
+
+/// Parameters for indexing operations (internal use)
+#[derive(Debug, Clone)]
+pub struct IndexingParams {
+    /// Path to the directory or file to index
+    pub path: PathBuf,
+    /// Human-readable name for the context
+    pub name: String,
+    /// Description of the context
+    pub description: String,
+    /// Whether this context should be persistent
+    pub persistent: bool,
+    /// Optional patterns to include during indexing
+    pub include_patterns: Option<Vec<String>>,
+    /// Optional patterns to exclude during indexing
+    pub exclude_patterns: Option<Vec<String>>,
+}
+
 use crate::client::SemanticContext;
 
 /// Type alias for context ID
@@ -52,6 +86,14 @@ pub struct KnowledgeContext {
     /// Original source path if created from a directory
     pub source_path: Option<String>,
 
+    /// Include patterns used during indexing
+    #[serde(default)]
+    pub include_patterns: Vec<String>,
+
+    /// Exclude patterns used during indexing
+    #[serde(default)]
+    pub exclude_patterns: Vec<String>,
+
     /// Number of items in the context
     pub item_count: usize,
 }
@@ -64,6 +106,7 @@ impl KnowledgeContext {
         description: &str,
         persistent: bool,
         source_path: Option<String>,
+        patterns: (Vec<String>, Vec<String>),
         item_count: usize,
     ) -> Self {
         let now = Utc::now();
@@ -74,6 +117,8 @@ impl KnowledgeContext {
             created_at: now,
             updated_at: now,
             source_path,
+            include_patterns: patterns.0,
+            exclude_patterns: patterns.1,
             persistent,
             item_count,
         }
@@ -304,6 +349,8 @@ pub(crate) enum IndexingJob {
         name: String,
         description: String,
         persistent: bool,
+        include_patterns: Option<Vec<String>>,
+        exclude_patterns: Option<Vec<String>>,
     },
     Clear {
         id: Uuid,
diff --git a/docs/knowledge-management.md b/docs/knowledge-management.md
index 3c4cc084e..816e85a8b 100644
--- a/docs/knowledge-management.md
+++ b/docs/knowledge-management.md
@@ -25,19 +25,57 @@ Once enabled, you can use `/knowledge` commands within your chat session:
 
 Display all entries in your knowledge base with detailed information including creation dates, item counts, and persistence status.
 
-#### `/knowledge add <name> <path>`
+#### `/knowledge add <name> <path> [--include pattern] [--exclude pattern]`
 
 Add files or directories to your knowledge base. The system will recursively index all supported files in directories.
 
 `/knowledge add "project-docs" /path/to/documentation`
 `/knowledge add "config-files" /path/to/config.json`
 
-Supported file types:
+**Default Pattern Behavior**
 
-- Text files: .txt
-- Markdown: .md, .markdown
-- JSON: .json
+When you don't specify `--include` or `--exclude` patterns, the system uses your configured default patterns:
+
+- If no patterns are specified and no defaults are configured, all supported files are indexed
+- Default include patterns apply when no `--include` is specified
+- Default exclude patterns apply when no `--exclude` is specified
+- Explicit patterns always override defaults
+
+Example with defaults configured:
+```bash
+q settings knowledge.defaultIncludePatterns '["**/*.rs", "**/*.py"]'
+q settings knowledge.defaultExcludePatterns '["target/**", "__pycache__/**"]'
+
+# This will use the default patterns
+/knowledge add "my-project" /path/to/project
+
+# This will override defaults with explicit patterns
+/knowledge add "docs-only" /path/to/project --include "**/*.md"
+```
+
+**New: Pattern Filtering**
+
+You can now control which files are indexed using include and exclude patterns:
+
+`/knowledge add "rust-code" /path/to/project --include "*.rs" --exclude "target/**"`
+`/knowledge add "docs" /path/to/project --include "**/*.md" --include "**/*.txt" --exclude "node_modules/**"`
+
+Pattern examples:
+- `*.rs` - All Rust files in all directories recursively (equivalent to `**/*.rs`)
+- `**/*.py` - All Python files recursively
+- `target/**` - Everything in target directory
+- `node_modules/**` - Everything in node_modules directory
+
+Supported file types (expanded):
+
+- Text files: .txt, .log, .rtf, .tex, .rst
+- Markdown: .md, .markdown, .mdx (now supported!)
+- JSON: .json (now treated as text for better searchability)
+- Configuration: .ini, .conf, .cfg, .properties, .env
+- Data files: .csv, .tsv
+- Web formats: .svg (text-based)
 - Code files: .rs, .py, .js, .jsx, .ts, .tsx, .java, .c, .cpp, .h, .hpp, .go, .rb, .php, .swift, .kt, .kts, .cs, .sh, .bash, .zsh, .html, .htm, .xml, .css, .scss, .sass, .less, .sql, .yaml, .yml, .toml
+- Special files: Dockerfile, Makefile, LICENSE, CHANGELOG, README (files without extensions)
 
 > Important: Unsupported files are indexed without text content extraction.
 
@@ -50,7 +88,7 @@ Remove entries from your knowledge base. You can remove by name, path, or contex
 
 #### `/knowledge update <path>`
 
-Update an existing knowledge base entry with new content from the specified path.
+Update an existing knowledge base entry with new content from the specified path. The original include/exclude patterns are preserved during updates.
 
 `/knowledge update /path/to/updated/project`
 
@@ -73,17 +111,28 @@ Cancel background operations. You can cancel a specific operation by ID or all o
 `/knowledge cancel abc12345 # Cancel specific operation`
 `/knowledge cancel all # Cancel all operations`
 
+## Configuration
+
+Configure knowledge base behavior:
+
+`q settings knowledge.maxFiles 10000` # Maximum files per knowledge base
+`q settings knowledge.chunkSize 1024` # Text chunk size for processing
+`q settings knowledge.chunkOverlap 256` # Overlap between chunks
+`q settings knowledge.defaultIncludePatterns '["**/*.rs", "**/*.md"]'` # Default include patterns
+`q settings knowledge.defaultExcludePatterns '["target/**", "node_modules/**"]'` # Default exclude patterns
+
 ## How It Works
 
 #### Indexing Process
 
 When you add content to the knowledge base:
 
-1. File Discovery: The system recursively scans directories for supported file types
-2. Content Extraction: Text content is extracted from each supported file
-3. Chunking: Large files are split into smaller, searchable chunks
-4. Background Processing: Indexing happens asynchronously in the background
-5. Semantic Embedding: Content is processed for semantic search capabilities
+1. **Pattern Filtering**: Files are filtered based on include/exclude patterns (if specified)
+2. **File Discovery**: The system recursively scans directories for supported file types
+3. **Content Extraction**: Text content is extracted from each supported file
+4. **Chunking**: Large files are split into smaller, searchable chunks
+5. **Background Processing**: Indexing happens asynchronously in the background
+6. **Semantic Embedding**: Content is processed for semantic search capabilities
 
 #### Search Capabilities
 
@@ -97,6 +146,7 @@ The knowledge base uses semantic search, which means:
 
 - Persistent contexts: Survive across chat sessions and CLI restarts
 - Context persistence is determined automatically based on usage patterns
+- Include/exclude patterns are stored with each context and reused during updates
 
 #### Best Practices
 
@@ -104,6 +154,7 @@ Organizing Your Knowledge Base
 
 - Use descriptive names when adding contexts: "api-documentation" instead of "docs"
 - Group related files in directories before adding them
+- Use include/exclude patterns to focus on relevant files
 - Regularly review and update outdated contexts
 
 #### Effective Searching
@@ -116,23 +167,31 @@ Organizing Your Knowledge Base
 #### Managing Large Projects
 
 - Add project directories rather than individual files when possible
+- Use include/exclude patterns to avoid indexing build artifacts: `--exclude "target/**" --exclude "node_modules/**"`
 - Use /knowledge status to monitor indexing progress for large directories
 - Consider breaking very large projects into logical sub-directories
 
+#### Pattern Filtering Best Practices
+
+- **Be specific**: Use precise patterns to avoid over-inclusion
+- **Exclude build artifacts**: Always exclude directories like `target/**`, `node_modules/**`, `.git/**`
+- **Include relevant extensions**: Focus on file types you actually need to search
+- **Test patterns**: Verify patterns match expected files before large indexing operations
+
 ## Limitations
 
 #### File Type Support
 
-- .mdx files are not currently supported for content extraction
 - Binary files are ignored during indexing
-- Very large files may be chunked, potentially splitting related content.
+- Very large files may be chunked, potentially splitting related content
+- Some specialized file formats may not extract content optimally
 
 #### Performance Considerations
 
 - Large directories may take significant time to index
 - Background operations are limited by concurrent processing limits
-- Search performance may vary based on knowledge base size
-- Currently there’s a hard limit of 5k files per knowledge base (getting removed soon as on Jul 12th, 2025).
+- Search performance may vary based on knowledge base size and embedding engine
+- Pattern filtering happens during file walking, improving performance for large directories
 
 #### Storage and Persistence
 
@@ -146,24 +205,37 @@ Organizing Your Knowledge Base
 
 If your files aren't appearing in search results:
 
-1. Check file types: Ensure your files have supported extensions
-2. Monitor status: Use /knowledge status to check if indexing is still in progress
-3. Verify paths: Ensure the paths you added actually exist and are accessible
-4. Check for errors: Look for error messages in the CLI output
+1. **Check patterns**: Ensure your include patterns match the files you want
+2. **Verify exclude patterns**: Make sure exclude patterns aren't filtering out desired files
+3. **Check file types**: Ensure your files have supported extensions
+4. **Monitor status**: Use /knowledge status to check if indexing is still in progress
+5. **Verify paths**: Ensure the paths you added actually exist and are accessible
+6. **Check for errors**: Look for error messages in the CLI output
 
 #### Search Not Finding Expected Results
 
 If searches aren't returning expected results:
 
-1. Wait for indexing: Use /knowledge status to ensure indexing is complete
-2. Try different queries: Use various phrasings and keywords
-3. Verify content: Use /knowledge show to confirm your content was added
-4. Check file types: Unsupported file types won't have searchable content
+1. **Wait for indexing**: Use /knowledge status to ensure indexing is complete
+2. **Try different queries**: Use various phrasings and keywords
+3. **Verify content**: Use /knowledge show to confirm your content was added
+4. **Check file types**: Unsupported file types won't have searchable content
 
 #### Performance Issues
 
 If operations are slow:
 
-1. Check queue status: Use /knowledge status to see operation queue
-2. Cancel if needed: Use /knowledge cancel to stop problematic operations
-3. Add smaller chunks: Consider adding subdirectories instead of entire large projects
+1. **Check queue status**: Use /knowledge status to see operation queue
+2. **Cancel if needed**: Use /knowledge cancel to stop problematic operations
+3. **Add smaller chunks**: Consider adding subdirectories instead of entire large projects
+4. **Use better patterns**: Exclude unnecessary files with exclude patterns
+5. **Adjust settings**: Consider lowering maxFiles or chunkSize for better performance
+
+#### Pattern Issues
+
+If patterns aren't working as expected:
+
+1. **Test patterns**: Use simple patterns first, then add complexity
+2. **Check syntax**: Ensure glob patterns use correct syntax (`**` for recursive, `*` for single level)
+3. **Verify paths**: Make sure patterns match actual file paths in your project
+4. **Use absolute patterns**: Consider using full paths in patterns for precision