sigmap 7.29.0 → 7.30.0

package/CHANGELOG.md

@@ -10,6 +10,19 @@ Format: [Semantic Versioning](https://semver.org/)
 
 ---
 
+## [7.30.0] — 2026-06-23
+
+Minor release — **v8.0 E2 + E4 (the "Pivot"):** completes v8.0 by repositioning every public surface to the chosen framing — *"the deterministic, verifiable grounding layer for AI code work"* — and framing coding agents as **consumers, not competitors**. The Evidence Pack code (E1/E3/D3 + `mcp install`) already shipped in 7.27–7.29; this is the positioning half. Docs/strings only — no runtime behaviour change, zero new dependencies.
+
+### Added
+- **Agent recipes (#389):** new README "Agent recipes" section with copy-paste setup for Claude Code, Cursor, Cline, Continue, Aider, OpenHands, and Codex CLI — each via `sigmap mcp install <client>` or a deterministic Evidence Pack, positioning agents as consumers of SigMap's map.
+- **Surface docs for shipped commands (#389):** README now documents `sigmap evidence` (deterministic Evidence Pack JSON/Markdown) and `sigmap doctor` (setup diagnostics), which shipped in code but were undocumented.
+- **Repositioning gate (#389):** `test/integration/repositioning.test.js` makes the pivot non-regressable — asserts the grounding-layer framing on README/`llms.txt`/docs `<title>`, recipes for every named agent, and the documented commands.
+
+### Changed
+- **E2 repositioning (#389):** README tagline, "What is SigMap?", "Why SigMap?" (token reduction demoted to proof) and the compare table; `docs/index.html` title/meta/keywords/JSON-LD + hero (and the stale `softwareVersion` 5.8.0 → current); `llms.txt`/`llms-full.txt` regenerated from `scripts/llms-manual.mjs`; the per-project adapter tagline in `src/format/llms-txt.js` (bundle rebuilt, reproducible); `docs/_config.yml`. The literal `context-engine` remains only inside the published JetBrains plugin URL slug.
+- **Structure guards updated (#389):** `readme-structure.test.js` tagline/compare-table assertions moved to the new copy; `version-json.test.js` now derives the docs `softwareVersion` from `version.json` instead of a hardcoded stale value.
+
 ## [7.29.0] — 2026-06-23
 
 Minor release — **v8.0 E4:** one-command, per-client MCP install so a cold user reaches a working MCP setup fast (the v8.0 <5-minute-quickstart exit gate).

package/README.md

@@ -4,7 +4,7 @@
 
 # ⚡ SigMap
 
-**SigMap finds the right files before your AI answers.**
+**SigMap is the deterministic, verifiable grounding layer for AI code work.**
 
 [![npm version](https://img.shields.io/npm/v/sigmap?color=7c6af7&label=latest&logo=npm)](https://www.npmjs.com/package/sigmap)
 [![npm downloads](https://img.shields.io/npm/dt/sigmap?color=22c55e&label=downloads&logo=npm)](https://www.npmjs.com/package/sigmap)
@@ -35,7 +35,9 @@ Zero config. Zero dependencies. Under 10 seconds.
 
 ## What is SigMap?
 
-SigMap extracts function and class signatures from your codebase and feeds the right files — not the whole repo — to your AI.
+SigMap builds a **deterministic, auditable signature-and-evidence map** of your codebase — no LLM calls, no embeddings, byte-stable output — so AI agents, CI, and reviewers can *trust and verify* which files and symbols are real before acting. Same repo in, same map out, every time.
+
+That map is exactly what agentic grep is worst at: reproducible, auditable context an agent can consume without a copy-paste, and a grounding check that proves an AI answer is anchored to real signatures and line numbers. Token reduction comes for free — but trust is the point.
 
 **Model-agnostic.** Works with:
 - **Cloud LLMs:** Claude, GPT-4, Copilot, Gemini
@@ -48,6 +50,12 @@ SigMap extracts function and class signatures from your codebase and feeds the r
 
 ## Why SigMap?
 
+**Deterministic and verifiable — the two things an agentic-grep loop can't give you:**
+- **Deterministic** — no LLM calls, no agent loop; the same repo always produces a byte-identical map you can diff, cache, and gate in CI.
+- **Auditable & grounded** — every file and symbol traces to a real line anchor; `sigmap verify-ai-output` flags any AI claim that isn't.
+- **Zero dependencies** — `npx sigmap` on any machine; no embeddings, no vector DB, no hosted service, fully offline.
+
+**Proof it pays off** (full benchmark below):
 <!--SM:whyMetrics-->
 - **75.6% hit@5** — right file found in top 5 results (vs 13.6% baseline)
 - **97.0% token reduction** — average across 21 real repos
@@ -58,7 +66,6 @@ SigMap extracts function and class signatures from your codebase and feeds the r
 - **No vendor lock-in** — works with any AI assistant or local LLM
 - **No API costs** — use local models (Ollama, llama.cpp, vLLM) with zero token fees
 - **Full privacy** — keep your code and context on your machine
-- **Zero npm dependencies** — `npx sigmap` on any machine
 
 ---
 
@@ -66,9 +73,9 @@ SigMap extracts function and class signatures from your codebase and feeds the r
 
 | Without SigMap | With SigMap |
 |---|---|
-| ❌ Guessing which files are relevant | ✅ Right file in context — <!--SM:hitWhole-->76%<!--/SM:hitWhole--> of the time |
-| ❌ Sending the full repo to your AI | ✅ Minimal context — only what matters |
-| ❌ Embeddings / vector DB required | ✅ Grounded answers, no infra needed |
+| ❌ Non-reproducible agent guesses | ✅ Deterministic map — same input, same output, every time |
+| ❌ "Trust me" AI answers | ✅ Grounded — right file in context <!--SM:hitWhole-->76%<!--/SM:hitWhole--> of the time, every symbol on a real line anchor |
+| ❌ Embeddings / vector DB required | ✅ Zero deps, no infra, fully offline |
 
 ---
 
@@ -91,7 +98,7 @@ Ask → Rank → Context → Validate → Judge → Learn
 
 <!--SM:benchmarkBlock-->
 ```
-Benchmark : sigmap-v7.29-main (21 repositories, including R language)
+Benchmark : sigmap-v7.30-main (21 repositories, including R language)
 Date      : 2026-06-23
 
 Hit@5          : 75.6%   (baseline 13.6%  — 5.6× lift)
@@ -216,6 +223,47 @@ sigmap create "<task>"             # run the whole pipeline: scaffold → verify
 
 ---
 
+## Evidence Pack & diagnostics
+
+The **Evidence Pack** is the consumable, machine-readable replacement for "paste this into your prompt" — a deterministic JSON artifact (with a Markdown handoff mode) that an agent or CI step reads directly, with zero copy-paste:
+
+```bash
+sigmap evidence "how does auth work"            # → .context/evidence-pack.json (deterministic, byte-stable)
+sigmap evidence "how does auth work" --markdown # Markdown handoff to stdout
+sigmap doctor                                   # diagnose config, index, freshness, coverage, MCP wiring — with fixes
+```
+
+Each pack carries the ranked files, the symbols and line anchors that justify them, the token budget, the dropped files (and why), and the grounding summary — so a consumer can trust and audit the context instead of guessing.
+
+---
+
+## Agent recipes
+
+SigMap treats coding agents as **consumers, not competitors**: it hands them a deterministic, auditable map the agent can read on demand. Wire any of them up once, then let the agent pull context or consume an Evidence Pack.
+
+| Agent | One-time setup | How it consumes SigMap |
+|---|---|---|
+| **Claude Code** | `sigmap mcp install claude` | 17 MCP tools (`search_signatures`, `get_lines`, `get_diff_context`…) |
+| **Cursor** | `sigmap mcp install cursor` | MCP tools, plus the `cursor` adapter writes `.cursorrules` |
+| **Cline** | `sigmap mcp install cursor` | Reads `.cursorrules`; same MCP server |
+| **Continue** | `sigmap mcp install vscode` | MCP tools inside the Continue extension |
+| **Aider** | `sigmap --adapter openai` | Reads `.github/openai-context.md` before a session |
+| **OpenHands** | `sigmap evidence "<task>"` | Consumes `.context/evidence-pack.json` directly |
+| **Codex CLI** | `sigmap mcp install codex` | MCP tools, plus the `codex` adapter writes `AGENTS.md` |
+
+```bash
+# Pattern 1 — give the agent live, on-demand access (MCP)
+sigmap mcp install claude        # one of: claude|cursor|windsurf|vscode|zed|codex|gemini|opencode|mcp
+                                 # add --global for a user-level install
+
+# Pattern 2 — hand the agent a deterministic Evidence Pack (no MCP, no copy-paste)
+sigmap evidence "implement rate limiting" --markdown   # or read .context/evidence-pack.json
+```
+
+See [`sigmap mcp list`](https://sigmap.io/guide/cli.html) for every supported client.
+
+---
+
 ## Try it
 
 ```bash

package/gen-context.js

@@ -9837,7 +9837,7 @@ __factories["./src/format/llms-txt"] = function(module, exports) {
 
     const lines = [
       '# SigMap Context Index',
-      `> Generated by SigMap v${sigmapVersion} — zero-dependency AI context engine`,
+      `> Generated by SigMap v${sigmapVersion} — the deterministic, verifiable grounding layer for AI code work`,
       '',
       '## Project',
       `- Name: ${name}`,
@@ -12695,7 +12695,7 @@ __factories["./src/mcp/server"] = function(module, exports) {
 
   const SERVER_INFO = {
     name: 'sigmap',
-    version: '7.29.0',
+    version: '7.30.0',
     description: 'SigMap MCP server — code signatures on demand',
   };
 
@@ -16524,7 +16524,7 @@ function __tryGit(args, opts = {}) {
   catch (_) { return ''; }
 }
 
-const VERSION = '7.29.0';
+const VERSION = '7.30.0';
 const MARKER = '\n\n## Auto-generated signatures\n<!-- Updated by gen-context.js -->\n';
 
 function requireSourceOrBundled(key) {
@@ -19091,7 +19091,7 @@ function main() {
     }
 
     const shareText = [
-      'Generated with SigMap — zero-dependency AI context engine',
+      'Generated with SigMap — the deterministic, verifiable grounding layer for AI code work',
       `${reduction}% fewer tokens · ${hitAt5}% retrieval accuracy · 6× better results`,
       'https://sigmap.io',
     ].join('\n');

package/llms-full.txt

@@ -1,21 +1,23 @@
 # SigMap — Complete LLM Reference
 
-> Zero-dependency context engine that stops AI coding agents from guessing in your codebase.
-> Grounded context. Verified answers. No embeddings, no vector DB, fully offline.
-
-SigMap is a zero-dependency AI context engine. It extracts function and class
-signatures from a codebase and uses TF-IDF ranking to feed an AI assistant only
-the files relevant to the task — cutting tokens ~97% while keeping answers
-grounded. Deterministic, offline, no embeddings or vector database. Works with
-Claude, Cursor, GitHub Copilot, Aider, Windsurf, local LLMs, and MCP.
-
-# Version: 7.29.0 | Benchmark: sigmap-v7.29-main (2026-06-23)
+> The deterministic, verifiable grounding layer for AI code work.
+> A reproducible signature-and-evidence map that agents, CI, and reviewers can trust and audit. No embeddings, no vector DB, fully offline.
+
+SigMap is the deterministic, verifiable grounding layer for AI code work. It
+extracts function and class signatures from a codebase and builds a byte-stable
+signature-and-evidence map that agents, CI, and reviewers can trust and audit —
+proving which files and symbols are real before acting. Deterministic TF-IDF
+ranking keeps the relevant context in scope (cutting tokens ~97% as a side
+effect), with no LLM calls, embeddings, or vector database. Works with Claude,
+Cursor, GitHub Copilot, Aider, Windsurf, local LLMs, and MCP.
+
+# Version: 7.30.0 | Benchmark: sigmap-v7.30-main (2026-06-23)
 # Source: auto-generated from package.json, version.json, benchmarks/latest.json, src/mcp/tools.js, src/config/defaults.js
 # Regenerate: npm run generate:llms   |   Validate: npm run validate:llms
 
 ---
 
-## Core metrics (benchmark: sigmap-v7.29-main, 2026-06-23)
+## Core metrics (benchmark: sigmap-v7.30-main, 2026-06-23)
 
 | Metric | Without SigMap | With SigMap |
 |--------|----------------|-------------|

package/llms.txt

@@ -1,15 +1,17 @@
 # SigMap
 
-> Zero-dependency context engine that stops AI coding agents from guessing in your codebase.
-> Grounded context. Verified answers. No embeddings, no vector DB, fully offline.
-
-SigMap is a zero-dependency AI context engine. It extracts function and class
-signatures from a codebase and uses TF-IDF ranking to feed an AI assistant only
-the files relevant to the task — cutting tokens ~97% while keeping answers
-grounded. Deterministic, offline, no embeddings or vector database. Works with
-Claude, Cursor, GitHub Copilot, Aider, Windsurf, local LLMs, and MCP.
-
-# Version: 7.29.0 | Benchmark: sigmap-v7.29-main (2026-06-23)
+> The deterministic, verifiable grounding layer for AI code work.
+> A reproducible signature-and-evidence map that agents, CI, and reviewers can trust and audit. No embeddings, no vector DB, fully offline.
+
+SigMap is the deterministic, verifiable grounding layer for AI code work. It
+extracts function and class signatures from a codebase and builds a byte-stable
+signature-and-evidence map that agents, CI, and reviewers can trust and audit —
+proving which files and symbols are real before acting. Deterministic TF-IDF
+ranking keeps the relevant context in scope (cutting tokens ~97% as a side
+effect), with no LLM calls, embeddings, or vector database. Works with Claude,
+Cursor, GitHub Copilot, Aider, Windsurf, local LLMs, and MCP.
+
+# Version: 7.30.0 | Benchmark: sigmap-v7.30-main (2026-06-23)
 # Source: auto-generated from package.json, version.json, benchmarks/latest.json, src/mcp/tools.js, src/config/defaults.js
 # Regenerate: npm run generate:llms   |   Validate: npm run validate:llms
 
@@ -21,7 +23,7 @@ Claude, Cursor, GitHub Copilot, Aider, Windsurf, local LLMs, and MCP.
 - No blast-radius awareness before editing a hub file — `--impact` shows every file a change touches.
 - Pasted stack traces, CI logs, and JSON bloat the prompt — `squeeze` minimizes them and enriches the top frame from the symbol index.
 
-## Core metrics (benchmark: sigmap-v7.29-main, 2026-06-23)
+## Core metrics (benchmark: sigmap-v7.30-main, 2026-06-23)
 
 - hit@5 retrieval: 75.6% vs 13.6% random baseline (5.6× lift)
 - Token reduction: 97.0% average across benchmark repos
@@ -52,5 +54,6 @@ npx sigmap --mcp                    # start the MCP server over stdio
 - [Benchmark dataset (Zenodo)](https://doi.org/10.5281/zenodo.19898842)
 - [Full LLM reference](https://sigmap.io/llms-full.txt)
 
-SigMap — grounded AI coding context. The lightweight, deterministic alternative
-to embeddings for feeding the right code to AI assistants.
+SigMap — the deterministic, verifiable grounding layer for AI code work. The
+reproducible signature-and-evidence map agents, CI, and reviewers can audit,
+which agentic grep cannot produce.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap",
-  "version": "7.29.0",
+  "version": "7.30.0",
   "description": "97% token reduction for AI coding. Extracts function & class signatures with TF-IDF ranking to feed only the right files to Claude, Cursor, Copilot, Aider, Windsurf, local LLMs & MCP. Zero dependencies, runs offline via npx.",
   "main": "packages/core/index.js",
   "exports": {

package/packages/cli/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap-cli",
-  "version": "7.29.0",
+  "version": "7.30.0",
   "description": "SigMap CLI wrapper — thin adapter for programmatic CLI invocation",
   "main": "index.js",
   "keywords": [

package/packages/core/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sigmap-core",
-  "version": "7.29.0",
+  "version": "7.30.0",
   "description": "SigMap core library — zero-dependency code signature extraction, retrieval, and security scanning",
   "main": "index.js",
   "keywords": [

package/src/format/llms-txt.js

@@ -29,7 +29,7 @@ function format(context, cwd, writtenFiles, sigmapVersion) {
 
   const lines = [
     '# SigMap Context Index',
-    `> Generated by SigMap v${sigmapVersion} — zero-dependency AI context engine`,
+    `> Generated by SigMap v${sigmapVersion} — the deterministic, verifiable grounding layer for AI code work`,
     '',
     '## Project',
     `- Name: ${name}`,

package/src/mcp/server.js

@@ -18,7 +18,7 @@ const { readContext, searchSignatures, getMap, createCheckpoint, getRouting, exp
 
 const SERVER_INFO = {
   name: 'sigmap',
-  version: '7.29.0',
+  version: '7.30.0',
   description: 'SigMap MCP server — code signatures on demand',
 };