memtrace 0.4.60 → 0.5.1

package/README.md

@@ -24,7 +24,7 @@ Memtrace gives coding agents something they've never had: **structural memory**.
 Index once. Every agent query after that resolves through graph traversal — callers, callees, implementations, imports, blast radius, temporal evolution — in milliseconds, with zero token waste.
 
 ```bash
-npm install -g memtrace    # binary + 17 skills + MCP server — one command
+npm install -g memtrace    # binary + 18 skills + MCP server — one command
 memtrace start             # launches the graph database and auto-indexes the current project
 ```
 
@@ -175,7 +175,7 @@ Memtrace exposes a full structural toolkit via the [Model Context Protocol](http
 </tr>
 </table>
 
-## 17 Agent Skills
+## 18 Agent Skills
 
 Memtrace ships skills that teach agents *how* to use the graph. They fire automatically based on what you ask — no prompt engineering required.
 
@@ -191,12 +191,13 @@ Memtrace ships skills that teach agents *how* to use the graph. They fire automa
 | **Index** | `memtrace-index` | _"index this project"_, _"parse this codebase"_ |
 | **Co-change** | `memtrace-cochange` | _"what else changes with this"_, _"hidden coupling"_ |
 
-Plus **8 workflow skills** that chain multiple tools with decision logic:
+Plus **9 workflow skills** that chain multiple tools with decision logic:
 
 | Skill | You say... |
 |:------|:-----------|
 | `memtrace-codebase-exploration` | _"I'm new to this project"_, _"give me an overview"_ |
 | `memtrace-change-impact-analysis` | _"what will break if I refactor this"_ |
+| `memtrace-code-review` | _"review this PR"_, _"post Memtrace review"_ |
 | `memtrace-incident-investigation` | _"something broke"_, _"root cause analysis"_ |
 | `memtrace-refactoring-guide` | _"help me refactor"_, _"clean up tech debt"_ |
 | `memtrace-first` | _"use Memtrace before file search"_ |
@@ -241,7 +242,7 @@ Uses **Structural Significance Budgeting** to surface the minimum set of changes
 
 ### Claude Code + Claude Desktop
 
-`npm install -g memtrace` handles everything automatically — binary, 17 skills, MCP server, plugin, and marketplace all register in one command for both Claude Code and Claude Desktop.
+`npm install -g memtrace` handles everything automatically — binary, 18 skills, MCP server, plugin, and marketplace all register in one command for both Claude Code and Claude Desktop.
 
 For manual setup:
 
@@ -257,7 +258,7 @@ Cursor **v2.4+** supports Agent Skills natively, and `npm install -g memtrace` h
 
 What the installer writes:
 - **MCP server** → `~/.cursor/mcp.json` (global — works in every project you open)
-- **17 skills/workflows** → `~/.cursor/skills/memtrace-*/SKILL.md`
+- **18 skills/workflows** → `~/.cursor/skills/memtrace-*/SKILL.md`
 
 For a **project-local** install (so the skills travel with your repo and teammates get them on clone), run inside the project:
 

package/installer/skills/workflows/memtrace-code-review.md

@@ -0,0 +1,42 @@
+---
+name: memtrace-code-review
+description: "Always use when the user asks to review a GitHub pull request, run Memtrace code review, post Memtrace review comments, create a PR with a review step, or publish local graph-backed review findings to GitHub. Prefer the review_github_pr MCP tool over manual diff inspection."
+when_to_use: "Triggered by 'review this PR', 'code review this pull request', 'post Memtrace review', 'run memtrace code-review', 'create PR with review', 'GitHub PR review', 'publish review comments', or any request to use Memtrace as a GitHub code reviewer. Use after code is pushed as a PR, and call review_github_pr instead of manually grepping diffs."
+paths: "**/*.{py,js,jsx,ts,tsx,mjs,cjs,rs,go,java,rb,c,cc,cpp,cxx,h,hpp,hh,cs,php,swift,kt,kts,scala,clj,cljs,ex,exs,erl,hrl,ml,mli,fs,fsx,r,jl,lua,dart,m,mm,asm,sql,gql,graphql,proto,sh,bash,zsh,fish,vue,svelte}"
+allowed-tools:
+  - mcp__memtrace__review_github_pr
+  - mcp__memtrace__find_code_review_issues
+  - mcp__memtrace__list_indexed_repositories
+user-invocable: true
+---
+
+## Overview
+
+Use Memtrace's local-first PR review workflow. The agent should call the `review_github_pr` MCP tool so review runs against the developer's local indexed graph, AST detectors, YAML rule pack, and review policy. GitHub is used for PR context and optional comment publication; source code analysis stays local.
+
+## Default Flow
+
+1. If the user gives a GitHub PR URL and asks to inspect or review it, call `review_github_pr` with `post: false`.
+2. If the user explicitly asks to publish, post comments, or complete the PR review, call `review_github_pr` with `post: true`.
+3. Use `graphMode: "strict"` by default. Use `graphMode: "off"` only when the user asks to benchmark non-graph behavior or the local graph is unavailable.
+4. Default to `minSeverity: "high"` and `maxComments: 5` when posting. For previews, `maxComments: 10` is acceptable.
+5. Pass `repoRoot` when the PR checkout is not the current working directory. Pass `repoId` when the indexed repository id is known.
+
+## Guardrails
+
+- Do not start with generic grep, rg, or manual diff review when `review_github_pr` is available.
+- Do not post comments unless the user explicitly requested publication.
+- Do not create benchmark-specific or PR-specific findings. The review must come from general Memtrace detectors, graph evidence, and policy ranking.
+- If the tool reports missing auth, tell the user to run `memtrace auth login`.
+- If the tool reports missing GitHub App installation, tell the user to install Memtrace Code Reviewer on that repository.
+- If the tool reports missing local graph context, tell the user to run `memtrace index .` at the workspace root.
+
+## Output
+
+For previews, summarize:
+- PR URL and repository
+- Graph state
+- Number of candidate comments
+- File, line, severity, and message for each finding
+
+For posted reviews, report the PR URL and number of comments posted.

package/lib/skill-metadata.js

@@ -266,6 +266,11 @@ const DEFAULT_TRIGGERS = {
             "Triggered by 'tour of this codebase', 'how is this organized', 'where do I start', 'onboard me', 'explain the architecture', 'main flows', 'main modules', 'key classes', 'I just cloned this'.",
         paths: CODE_PATHS,
     },
+    "memtrace-code-review": {
+        whenToUse:
+            "Triggered by 'review this PR', 'code review this pull request', 'post Memtrace review', 'run memtrace code-review', 'create PR with review', 'GitHub PR review', 'publish review comments', or any request to use Memtrace as a GitHub code reviewer. Use after code is pushed as a PR and call review_github_pr instead of manually grepping diffs.",
+        paths: CODE_PATHS,
+    },
     "memtrace-continuous-memory": {
         whenToUse:
             "Triggered by 'watch this folder', 'live indexing', 'incremental memtrace', 'always-on memory', 'keep up to date', 'pick up my saves', 'reflect uncommitted changes', 'why isn't memtrace seeing my edits'.",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "memtrace",
-  "version": "0.4.60",
+  "version": "0.5.1",
   "description": "Code intelligence graph — MCP server + AI agent skills + visualization UI",
   "keywords": [
     "mcp",
@@ -39,11 +39,11 @@
     "fs-extra": "^11.0.0"
   },
   "optionalDependencies": {
-    "@memtrace/darwin-arm64": "0.4.60",
-    "@memtrace/linux-x64": "0.4.60",
-    "@memtrace/win32-x64": "0.4.60",
-    "@memtrace/linux-x64-noavx2": "0.4.60",
-    "@memtrace/win32-x64-noavx2": "0.4.60"
+    "@memtrace/darwin-arm64": "0.5.1",
+    "@memtrace/linux-x64": "0.5.1",
+    "@memtrace/win32-x64": "0.5.1",
+    "@memtrace/linux-x64-noavx2": "0.5.1",
+    "@memtrace/win32-x64-noavx2": "0.5.1"
   },
   "engines": {
     "node": ">=18"

package/skills/workflows/memtrace-code-review.md

@@ -0,0 +1,48 @@
+---
+name: memtrace-code-review
+description: "Always use when the user asks to review a GitHub pull request, run Memtrace code review, post Memtrace review comments, create a PR with a review step, or publish local graph-backed review findings to GitHub. Prefer the review_github_pr MCP tool over manual diff inspection."
+when_to_use: "Triggered by 'review this PR', 'code review this pull request', 'post Memtrace review', 'run memtrace code-review', 'create PR with review', 'GitHub PR review', 'publish review comments', or any request to use Memtrace as a GitHub code reviewer. Use after code is pushed as a PR, and call review_github_pr instead of manually grepping diffs."
+paths: "**/*.{py,js,jsx,ts,tsx,mjs,cjs,rs,go,java,rb,c,cc,cpp,cxx,h,hpp,hh,cs,php,swift,kt,kts,scala,clj,cljs,ex,exs,erl,hrl,ml,mli,fs,fsx,r,jl,lua,dart,m,mm,asm,sql,gql,graphql,proto,sh,bash,zsh,fish,vue,svelte}"
+allowed-tools:
+  - mcp__memtrace__review_github_pr
+  - mcp__memtrace__find_code_review_issues
+  - mcp__memtrace__list_indexed_repositories
+user-invocable: true
+---
+
+## Overview
+
+Use Memtrace's local-first PR review workflow. The agent should call the `review_github_pr` MCP tool so review runs against the developer's local indexed graph, AST detectors, YAML rule pack, and review policy. GitHub is used for PR context and optional comment publication; source code analysis stays local.
+
+## Default Flow
+
+1. If the user gives a GitHub PR URL and asks to inspect or review it, call `review_github_pr` with `post: false`.
+2. If the user explicitly asks to publish, post comments, or complete the PR review, call `review_github_pr` with `post: true`.
+3. Use `graphMode: "strict"` by default. Use `graphMode: "off"` only when the user asks to benchmark non-graph behavior or the local graph is unavailable.
+4. Default to `minSeverity: "high"` and `maxComments: 5` when posting. For previews, `maxComments: 10` is acceptable.
+5. Pass `repoRoot` when the PR checkout is not the current working directory. Pass `repoId` when the indexed repository id is known.
+
+## Example User Prompts
+
+- "Review this PR with Memtrace: https://github.com/OWNER/REPO/pull/123"
+- "Use Memtrace to review this pull request and post the findings: https://github.com/OWNER/REPO/pull/123"
+- "Create the PR, then run Memtrace code review and publish the review comments."
+
+## Guardrails
+
+- Do not start with generic grep, rg, or manual diff review when `review_github_pr` is available.
+- Do not post comments unless the user explicitly requested publication.
+- Do not create benchmark-specific or PR-specific findings. The review must come from general Memtrace detectors, graph evidence, and policy ranking.
+- If the tool reports missing auth, tell the user to run `memtrace auth login`.
+- If the tool reports missing GitHub App installation, tell the user to install Memtrace Code Reviewer on that repository.
+- If the tool reports missing local graph context, tell the user to run `memtrace index .` at the workspace root.
+
+## Output
+
+For previews, summarize:
+- PR URL and repository
+- Graph state
+- Number of candidate comments
+- File, line, severity, and message for each finding
+
+For posted reviews, report the PR URL and number of comments posted.

package/uninstall.js

@@ -48,6 +48,7 @@ const SKILL_NAMES = [
   // Workflows
   "memtrace-change-impact-analysis",
   "memtrace-codebase-exploration",
+  "memtrace-code-review",
   "memtrace-incident-investigation",
   "memtrace-refactoring-guide",
 ];