docs: update API documentation to reflect simplified read_email output

Changes:
- Updated TOOLS.md to reflect new read_email response structure
  - Removed headers and file_path from response examples
  - Added cc, references, in_reply_to fields
  - Updated "What gets extracted" section to clarify only key headers included
  - Updated read_thread example emails with new field structure
  - Removed file_path from error handling documentation

- Updated ARCHITECTURE.md to document parse_email changes
  - Added Cc field extraction to data flow example
  - Added threading headers extraction (References, In-Reply-To)
  - Updated process steps to reflect 10-step parsing flow
  - Clarified that output has no raw headers and no file path

This aligns documentation with the code simplification that removes
unnecessary data (full headers, file paths) and focuses on AI-relevant
information for email analysis.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Author: fatbobman

ARCHITECTURE.md

@@ -323,9 +323,11 @@ Mail groups related emails using `conversation_id`:
 3. Extract email content until plist XML marker
 4. Parse RFC 5322 format using `email` module
 5. Decode headers using `email.header.decode_header()`
-6. Extract `text/plain` body parts
-7. Decode body using charset from `Content-Type`
-8. Return structured JSON
+6. Extract key headers: Subject, From, To, Cc, Date
+7. Extract threading headers: References, In-Reply-To
+8. Extract `text/plain` body parts
+9. Decode body using charset from `Content-Type`
+10. Return structured JSON (no raw headers, no file path)
 
 **Supported Formats**:
 
@@ -407,7 +409,8 @@ Mail groups related emails using `conversation_id`:
 │ ┌──────────────────────────────────────────────────────┐   │
 │ │ parse_email.py                                       │   │
 │ │   → Read .emlx file                                  │   │
-│ │   → Extract Subject, From, To, Date                  │   │
+│ │   → Extract Subject, From, To, Cc, Date              │   │
+│ │   → Extract threading headers (References, In-Reply-To)│   │
 │ │   → Decode body text                                 │   │
 │ │   → Return JSON structure                            │   │
 │ └──────────────────────────────────────────────────────┘   │

README.md

@@ -7,7 +7,7 @@
 
 **Demo: See it in action** (Click to watch)
 
-[![Demo Video](https://github.com/user-attachments/assets/bd06230b-16ad-476d-8b73-4495071b2b6c)]
+![Demo Video](https://github.com/user-attachments/assets/b9229493-8bdf-4995-9ca8-d5de55ec7144)
 
 *Shows: Copy Message-ID from Mail → AI analyzes email content*
 

README_zh.md

@@ -7,7 +7,7 @@
 
 **演示: 实际使用效果** (点击观看)
 
-[![演示视频](https://github.com/user-attachments/assets/bd06230b-16ad-476d-8b73-4495071b2b6c)]
+![演示视频](https://github.com/user-attachments/assets/b9229493-8bdf-4995-9ca8-d5de55ec7144)
 
 *内容: 从 Mail 复制 Message-ID → AI 分析邮件内容*
 

TOOLS.md

@@ -182,18 +182,11 @@ Parse and read plain text content of a single email.
   "subject": "Project Update",
   "from": "Sender Name <sender@example.com>",
   "to": "Recipient Name <recipient@example.com>",
+  "cc": "cc@example.com",
   "date": "Wed, 25 Dec 2024 10:30:00 +0800",
-  "body_text": "Plain text email body...",
-  "headers": {
-    "Message-Id": "<abc@example.com>",
-    "Subject": "Project Update",
-    "From": "Sender Name <sender@example.com>",
-    "To": "Recipient Name <recipient@example.com>",
-    "Date": "Wed, 25 Dec 2024 10:30:00 +0800",
-    "Cc": "cc@example.com",
-    "X-Mailer": "Apple Mail (..."
-  },
-  "file_path": "/path/to/email.emlx"
+  "references": "<original@example.com> <prev@example.com>",
+  "in_reply_to": "<prev@example.com>",
+  "body_text": "Plain text email body..."
 }
 ```
 
@@ -202,8 +195,7 @@ Parse and read plain text content of a single email.
 ```json
 {
   "success": false,
-  "error": "Parse failed: ...",
-  "file_path": "/path/to/email.emlx"
+  "error": "Parse failed: ..."
 }
 ```
 
@@ -219,12 +211,14 @@ Parse and read plain text content of a single email.
 **What gets extracted**:
 
 - ✅ Plain text body (`text/plain` parts only)
-- ✅ Headers (Subject, From, To, Date, Cc, etc.)
+- ✅ Key headers (Subject, From, To, Cc, Date)
+- ✅ Threading headers (References, In-Reply-To)
 - ✅ Decoded header values (e.g. `=?UTF-8?B?...?=`)
 - ❌ Attachments (skipped)
 - ❌ HTML content (skipped)
 - ❌ Embedded images (skipped)
 - ❌ Rich text/RTF (skipped)
+- ❌ Full raw headers (not included)
 
 **Email Parsing Process**:
 
@@ -280,32 +274,35 @@ Parse and read entire email thread.
       "subject": "Re: Project Update",
       "from": "Alice <alice@example.com>",
       "to": "Bob <bob@example.com>",
+      "cc": "",
       "date": "Mon, 23 Dec 2024 09:00:00 +0800",
-      "body_text": "Hi Bob, how's the project going?",
-      "headers": {...},
-      "file_path": "/path/to/msg1.emlx"
+      "references": "<original@example.com>",
+      "in_reply_to": "",
+      "body_text": "Hi Bob, how's the project going?"
     },
     {
       "success": true,
       "message_id": "<msg2@example.com>",
       "subject": "Re: Project Update",
       "from": "Bob <bob@example.com>",
       "to": "Alice <alice@example.com>",
+      "cc": "",
       "date": "Tue, 24 Dec 2024 10:30:00 +0800",
-      "body_text": "Making good progress...",
-      "headers": {...},
-      "file_path": "/path/to/msg2.emlx"
+      "references": "<original@example.com> <msg1@example.com>",
+      "in_reply_to": "<msg1@example.com>",
+      "body_text": "Making good progress..."
     },
     {
       "success": true,
       "message_id": "<msg3@example.com>",
       "subject": "Re: Project Update",
       "from": "Alice <alice@example.com>",
       "to": "Bob <bob@example.com>",
+      "cc": "",
       "date": "Wed, 25 Dec 2024 15:45:00 +0800",
-      "body_text": "Great to hear!",
-      "headers": {...},
-      "file_path": "/path/to/msg3.emlx"
+      "references": "<original@example.com> <msg1@example.com> <msg2@example.com>",
+      "in_reply_to": "<msg2@example.com>",
+      "body_text": "Great to hear!"
     }
   ]
 }
@@ -360,7 +357,6 @@ All tools follow this error response format:
 Additional fields may be included depending on the error:
 
 - `message_id` - When Message-ID is provided
-- `file_path` - When file operation fails
 - `possible_reasons` - Array of likely causes (for `get_email_path`)
 
 ### Common Errors