sigmap 2.7.0 → 2.8.0

package/CHANGELOG.md

@@ -6,7 +6,27 @@ Format: [Semantic Versioning](https://semver.org/)
 
 ---
 
-## [2.7.0] — upcoming · [#19](https://github.com/manojmallick/sigmap/issues/19) · branch: `feat/v2.7-ranking-optimization`
+## [2.8.0] — upcoming · [#21](https://github.com/manojmallick/sigmap/issues/21) · branch: `feat/v2.8-snippet-retrieval`
+
+### Planned additions
+- **Snippet extraction** — `src/retrieval/snippets.js`: extract relevant code blocks (functions, classes, methods) from ranked files
+- **Hybrid scoring** — combine file-level relevance with snippet-level relevance; snippets inherit file score + get their own local score
+- **`--query --snippets` CLI flag** — return top-k snippets (not full file sigs), with line numbers and context
+- **`query_context` MCP enhancement** — add `snippets: true` option; response includes snippet text + line ranges
+- **Smart context window** — include 2-3 lines before/after snippet for context
+- **Configuration** — `retrieval.snippets: { enabled: true, minLines: 3, maxSnippets: 5 }`
+- **`test/integration/snippets.test.js`** — 12 tests: snippet extraction, scoring, line number accuracy, context window
+
+### Go / No-go criteria
+- All tests green (21 extractor + all integration)
+- `--query "extract signatures" --snippets` returns 3-5 relevant snippets with correct line numbers
+- MCP `query_context` with `snippets: true` returns snippet text
+- Snippet relevance improves precision@3 by ≥10% over full-file retrieval
+- Performance: <150ms for 1000-file repos with snippets enabled
+
+---
+
+## [2.7.0] — 2026-04-05 · [#19](https://github.com/manojmallick/sigmap/issues/19)
 
 ### Planned additions
 - **Fine-tuned ranking weights** — optimize `exactToken`, `symbolMatch`, `prefixMatch`, `pathMatch`, and `recencyBoost` weights in `src/retrieval/ranker.js` based on benchmark-driven evaluation

package/README.md

@@ -110,17 +110,17 @@ AI agent session starts with full context
 
 ---
 
-## 🔭 What's next — v2.7 (in progress · [#19](https://github.com/manojmallick/sigmap/issues/19))
+## 🔭 What's next — v2.8 (in progress · [#21](https://github.com/manojmallick/sigmap/issues/21))
 
-### v2.7 — Ranking Optimization
+### v2.8 — Hybrid Depth: Snippet Retrieval
 
 | Feature | Description |
 |---|---|
-| **Fine-tuned ranking weights** | Optimize weights in `src/retrieval/ranker.js` for better precision based on benchmark evaluation |
-| **TF-IDF scoring** | Add term frequency-inverse document frequency scoring option for better semantic relevance |
-| **Weight presets** | `precision`, `balanced`, `recall` presets — configurable via `retrieval.preset` |
-| **Performance optimization** | <100ms query performance for 1000-file repos, optimized for large codebases (10K+ files) |
-| **Precision improvement** | Target ≥5% precision@5 improvement while maintaining hit@5 ≥ 0.80 |
+| **Snippet extraction** | Extract relevant code blocks (functions, classes, methods) from ranked files |
+| **Hybrid scoring** | Combine file-level + snippet-level relevance for precision |
+| **`--query --snippets`** | Return top-k snippets with line numbers and context (not full files) |
+| **MCP enhancement** | `query_context` with `snippets: true` returns snippet text + line ranges |
+| **Smart context** | Include 2-3 lines before/after each snippet for readability |
 
 ---
 | **`get_impact` MCP tool** | 9th MCP tool — `{ file, depth? }` → impacted files + signatures |

package/gen-context.js

@@ -4304,7 +4304,7 @@ const path = require('path');
 const os = require('os');
 const { execSync } = require('child_process');
 
-const VERSION = '2.7.0';
+const VERSION = '2.8.0';
 const MARKER = '\n\n## Auto-generated signatures\n<!-- Updated by gen-context.js -->\n';
 
 function requireSourceOrBundled(key) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap",
-  "version": "2.7.0",
+  "version": "2.8.0",
   "description": "Zero-dependency AI context engine — 97% token reduction. No npm install. Runs on Node 18+.",
   "main": "gen-context.js",
   "exports": {

package/packages/cli/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap-cli",
-  "version": "2.7.0",
+  "version": "2.8.0",
   "description": "SigMap CLI wrapper — thin adapter for programmatic CLI invocation",
   "main": "index.js",
   "keywords": [

package/packages/core/README.md

@@ -128,9 +128,9 @@ const health = score('/path/to/project');
 
 All existing CLI flags (`--generate`, `--watch`, `--mcp`, `--query`, `--analyze`, `--benchmark`, `--health`, …) are unchanged.
 
-## What's next — v2.7
+## What's next — v2.8
 
-v2.7 adds ranking optimization and fine-tuned weights for better precision in query-aware retrieval. TF-IDF scoring option, configurable weight presets (precision, balanced, recall), and performance optimization for large codebases. See [issue #19](https://github.com/manojmallick/sigmap/issues/19).
+v2.8 adds hybrid depth snippet retrieval — return not just relevant files, but the exact code snippets within those files. Includes `--query --snippets` CLI flag and `snippets: true` option for MCP `query_context`. See [issue #21](https://github.com/manojmallick/sigmap/issues/21).
 
 See the full [roadmap](https://manojmallick.github.io/sigmap/roadmap.html).
 

package/packages/core/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap-core",
-  "version": "2.7.0",
+  "version": "2.8.0",
   "description": "SigMap core library — zero-dependency code signature extraction, retrieval, and security scanning",
   "main": "index.js",
   "keywords": [

package/src/mcp/server.js

@@ -18,7 +18,7 @@ const { readContext, searchSignatures, getMap, createCheckpoint, getRouting, exp
 
 const SERVER_INFO = {
   name: 'sigmap',
-  version: '2.7.0',
+  version: '2.8.0',
   description: 'SigMap MCP server — code signatures on demand',
 };