Skip to content

Commit d60ae14

Browse files
thedotmackclaude
andcommitted
docs: update CHANGELOG.md for v10.5.0
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
1 parent 272391e commit d60ae14

File tree

1 file changed

+35
-18
lines changed

1 file changed

+35
-18
lines changed

CHANGELOG.md

Lines changed: 35 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -2,6 +2,41 @@
22

33
All notable changes to claude-mem.
44

5+
## [v10.5.0] - 2026-02-26
6+
7+
## Smart Explore: AST-Powered Code Navigation
8+
9+
This release introduces **Smart Explore**, a token-optimized structural code search system built on tree-sitter AST parsing. It applies the same progressive disclosure pattern used in human-readable code outlines — but programmatically, for AI agents.
10+
11+
### Why This Matters
12+
13+
The standard exploration cycle (Glob → Grep → Read) forces agents to consume entire files to understand code structure. A typical 800-line file costs ~12,000 tokens to read. Smart Explore replaces this with a 3-layer progressive disclosure workflow that delivers the same understanding at **6-12x lower token cost**.
14+
15+
### 3 New MCP Tools
16+
17+
- **`smart_search`** — Walks directories, parses all code files via tree-sitter, and returns ranked symbols with signatures and line numbers. Replaces the Glob → Grep discovery cycle in a single call (~2-6k tokens).
18+
- **`smart_outline`** — Returns the complete structural skeleton of a file: all functions, classes, methods, properties, imports (~1-2k tokens vs ~12k for a full Read).
19+
- **`smart_unfold`** — Expands a single symbol to its full source code including JSDoc, decorators, and implementation (~1-7k tokens).
20+
21+
### Token Economics
22+
23+
| Approach | Tokens | Savings |
24+
|----------|--------|---------|
25+
| smart_outline + smart_unfold | ~3,100 | 8x vs Read |
26+
| smart_search (cross-file) | ~2,000-6,000 | 6-12x vs Explore agent |
27+
| Read (full file) | ~12,000+ | baseline |
28+
| Explore agent | ~20,000-40,000 | baseline |
29+
30+
### Language Support
31+
32+
10 languages via tree-sitter grammars: TypeScript, JavaScript, Python, Rust, Go, Java, C, C++, Ruby, PHP.
33+
34+
### Other Changes
35+
36+
- Simplified hooks configuration
37+
- Removed legacy setup.sh script
38+
- Security fix: replaced `execSync` with `execFileSync` to prevent command injection in file path handling
39+
540
## [v10.4.4] - 2026-02-26
641

742
## Fix
@@ -1197,21 +1232,3 @@ Comprehensive test suite in a new PR, targeting **v8.6.0**
11971232

11981233
🤖 Generated with [Claude Code](https://claude.com/claude-code)
11991234

1200-
## [v8.5.5] - 2026-01-03
1201-
1202-
## Improved Error Handling and Logging
1203-
1204-
This patch release enhances error handling and logging across all worker services for better debugging and reliability.
1205-
1206-
### Changes
1207-
- **Enhanced Error Logging**: Improved error context across SessionStore, SearchManager, SDKAgent, GeminiAgent, and OpenRouterAgent
1208-
- **SearchManager**: Restored error handling for Chroma calls with improved logging
1209-
- **SessionStore**: Enhanced error logging throughout database operations
1210-
- **Bug Fix**: Fixed critical bug where `memory_session_id` could incorrectly equal `content_session_id`
1211-
- **Hooks**: Streamlined error handling and loading states for better maintainability
1212-
1213-
### Investigation Reports
1214-
- Added detailed analysis documents for generator failures and observation duplication regressions
1215-
1216-
**Full Changelog**: https://github.com/thedotmack/claude-mem/compare/v8.5.4...v8.5.5
1217-

0 commit comments

Comments
 (0)