mnemo-dev 0.5.2__tar.gz → 0.6.0__tar.gz

{mnemo_dev-0.5.2 → mnemo_dev-0.6.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mnemo-dev
-Version: 0.5.2
+Version: 0.6.0
 Summary: Persistent memory and repo map for AI coding assistants
 Author: Mnemo Contributors
 License: AGPL-3.0
@@ -59,18 +59,18 @@ Dynamic: license-file
 <p align="center">
   <a href="https://pypi.org/project/mnemo-dev/"><img src="https://img.shields.io/pypi/v/mnemo-dev?style=flat-square&color=blue" alt="PyPI" /></a>
   <a href="https://www.npmjs.com/package/@mnemo-dev/mcp"><img src="https://img.shields.io/npm/v/@mnemo-dev/mcp?style=flat-square&color=red" alt="npm" /></a>
-  <a href="#"><img src="https://img.shields.io/badge/tests-222%20passing-brightgreen?style=flat-square" alt="Tests" /></a>
+  <a href="#"><img src="https://img.shields.io/badge/tests-641%20passing-brightgreen?style=flat-square" alt="Tests" /></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-AGPL--3.0-purple?style=flat-square" alt="License" /></a>
   <a href="https://pypi.org/project/mnemo-dev/"><img src="https://img.shields.io/pypi/pyversions/mnemo-dev?style=flat-square" alt="Python" /></a>
   <a href="https://marketplace.visualstudio.com/items?itemName=Nikhil1057.mnemo-vscode"><img src="https://img.shields.io/badge/VS%20Code-extension-007ACC?style=flat-square&logo=visualstudiocode" alt="VS Code" /></a>
 </p>
 
 <p align="center">
-  <code>[100% R@5]</code> <code>[2ms search]</code> <code>[58 tools]</code> <code>[16 agent-facing]</code> <code>[11 lifecycle hooks]</code> <code>[0 external DBs]</code> <code>[222 tests]</code>
+  <code>[100% R@5]</code> <code>[2ms search]</code> <code>[58 tools]</code> <code>[17 agent-facing]</code> <code>[11 lifecycle hooks]</code> <code>[0 external DBs]</code> <code>[641 tests]</code>
 </p>
 
 <p align="center">
-  <a href="#install">Install</a> • <a href="#why-mnemo">Why</a> • <a href="#benchmarks">Benchmarks</a> • <a href="#how-it-works">How It Works</a> • <a href="#supported-clients">Clients</a> • <a href="#features">Features</a> • <a href="#mcp-tools-16-agent-facing">Tools</a> • <a href="#dashboard-ui">Dashboard</a> • <a href="#architecture">Architecture</a>
+  <a href="#install">Install</a> • <a href="#why-mnemo">Why</a> • <a href="#benchmarks">Benchmarks</a> • <a href="#how-it-works">How It Works</a> • <a href="#supported-clients">Clients</a> • <a href="#features">Features</a> • <a href="#mcp-tools-17-agent-facing">Tools</a> • <a href="#dashboard-ui">Dashboard</a> • <a href="#architecture">Architecture</a>
 </p>
 
 ---
@@ -89,7 +89,7 @@ Session 2: you ask the agent to add a new endpoint. It already knows your servic
 ```bash
 pip install mnemo-dev    # or: brew tap Mnemo-mcp/tap && brew install mnemo
 cd your-project
-mnemo init              # defaults to Amazon Q (or: --client kiro, cursor, claude-code)
+mnemo init --client kiro # or: amazonq, cursor, claude-code, copilot, generic
 ```
 
 ---
@@ -141,8 +141,7 @@ Then initialize:
 
 ```bash
 cd your-project
-mnemo init                      # defaults to Amazon Q
-mnemo init --client kiro        # or: cursor, claude-code, copilot, generic
+mnemo init --client kiro        # or: amazonq, cursor, claude-code, copilot, generic
 ```
 
 **That's it.** Your agent now has persistent memory, semantic search, and architectural understanding.
@@ -267,13 +266,13 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 ## How It Works
 
 ```
-┌─── INIT (one-time, ~7s for 300 files) ──────────────────────────┐
+┌─── INIT (one-time, ~7s for 300 files) ───────────────────────────┐
 │                                                                  │
 │  1. Scan: single os.walk pass across repo                        │
 │  2. Parse: tree-sitter AST (14 langs) + Roslyn (C#)              │
 │  3. Graph: LadybugDB — files, classes, methods, CALLS edges      │
 │  4. Scope: cross-file function call resolution                   │
-│  5. Cluster: Leiden community detection                          │
+│  5. Cluster: Louvain community detection                         │
 │  6. Index: ONNX vector embeddings for semantic search            │
 │  7. Detect: languages, services, key classes, frameworks         │
 │  8. Configure: MCP server + hooks for your AI client             │
@@ -281,10 +280,10 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── SESSION START (automatic via hooks) ─────────────────────────┐
+┌─── SESSION START (automatic via hooks) ──────────────────────────┐
 │                                                                  │
 │  mnemo_recall injects into agent context:                        │
-│    • Architectural decisions (permanent, never evicted)           │
+│    • Architectural decisions (permanent, never evicted)          │
 │    • Hot memories (scored by access × recency × importance)      │
 │    • Active plan + next task                                     │
 │    • Compact repo index (classes per service)                    │
@@ -297,7 +296,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
                               ▼
 ┌─── DURING SESSION (tools + freshness) ──────────────────────────┐
 │                                                                  │
-│  Agent has 16 MCP tools available:                               │
+│  Agent has 17 MCP tools available:                               │
 │    • mnemo_lookup → full service/class architecture              │
 │    • mnemo_search → semantic search (code, memory, APIs)         │
 │    • mnemo_impact → blast radius if X changes                    │
@@ -310,7 +309,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── SESSION END (auto-capture via stop hook) ────────────────────┐
+┌─── SESSION END (auto-capture via stop hook) ─────────────────────┐
 │                                                                  │
 │  • Detects learnings (bug fixes, discoveries)                    │
 │  • Records session decisions                                     │
@@ -319,11 +318,11 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── BETWEEN SESSIONS (decay + maintenance) ──────────────────────┐
+┌─── BETWEEN SESSIONS (decay + maintenance) ───────────────────────┐
 │                                                                  │
 │  Every 10th recall:                                              │
 │    • Retention scored: salience × exp(-0.01 × days) + access     │
-│    • Hot (≥0.5) → Warm (≥0.25) → Cold → Evicted                 │
+│    • Hot (≥0.5) → Warm (≥0.25) → Cold → Evicted                  │
 │    • Contradictions auto-superseded (sim > 0.9)                  │
 │    • Low-value pruning (cap: 200 active memories)                │
 │    • Graph synced (stale memory nodes removed)                   │
@@ -331,7 +330,6 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 │  Pinned forever: architecture, decision, preference              │
 │                                                                  │
 └──────────────────────────────────────────────────────────────────┘
- └─────────────────────────────────────────────────────────────┘
 ```
 
 ---
@@ -340,6 +338,8 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 
 ### 🧠 Memory System
 - **Categorized storage**: architecture, pattern, bug, preference, decision, todo
+- **Event-sourced decisions**: JSONL append-only event log, computed snapshots, supersede/redact operations
+- **Branch-scoped decisions**: decisions tagged per branch, filtered on recall
 - **Retention scoring**: access frequency × recency × importance (Ebbinghaus-inspired decay)
 - **Branch-aware**: memories tagged with git branch, filtered on recall
 - **Contradiction detection**: new facts auto-supersede old conflicting ones (threshold: 0.6)
@@ -360,7 +360,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 - **Knowledge graph**: files, folders, classes, methods, functions, projects, communities
 - **14 languages**: Python, JS/TS, C#, Go, Java, Rust, Ruby, PHP, C/C++, Kotlin, Swift, Scala
 - **Roslyn enrichment**: C# method signatures, implements, full AST detail
-- **Leiden community detection**: automatic functional clustering
+- **Louvain community detection**: automatic functional clustering
 - **CALLS edges**: scope-resolved function call graph with confidence scoring
 - **Impact analysis**: upstream/downstream blast radius (N-hop BFS)
 - **Incremental freshness**: graph + vector index auto-update within 30s of file changes
@@ -394,9 +394,45 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 
 ---
 
-## MCP Tools (58 total: 16 agent-facing + 42 specialized)
+## Workflow Skills & Orchestrator
 
-Mnemo exposes **16 consolidated agent-facing tools** via MCP — designed to cover every workflow in minimal tool calls. Under the hood, these route to **42 specialized internal tools** for granular operations.
+Mnemo includes **6 workflow skills** that guide agents through a structured SDLC pipeline, with **quality gates** enforcing standards between phases.
+
+### The Pipeline
+
+```
+investigate → plan → implement → verify → review → ship
+                                    ↑                ↑
+                              [tests_pass]    [tests_pass + plan_done + no_findings]
+```
+
+Each phase has a dedicated skill template with explicit instructions, shell commands for mnemo tools, and persistence hooks. The orchestrator tracks state in `.mnemo/autorun_state.json` and blocks advancement when gates fail.
+
+### Quality Gates
+
+| Gate | What it checks | Blocks |
+|------|---------------|--------|
+| `tests_pass` | Runs test suite (pytest/npm/maven/gradle) | review, ship |
+| `plan_done` | All plan tasks marked complete | ship |
+| `no_findings` | No unresolved critical findings in memory | ship |
+
+### Template Resolver System
+
+Skills use `{{RESOLVER}}` placeholders that get expanded per-host:
+
+| Resolver | Injects |
+|----------|---------|
+| `{{PREAMBLE}}` | Mandatory header — forces tool usage, DO NOT PROCEED pattern |
+| `{{LEARNINGS}}` | Past learnings relevant to current skill |
+| `{{CONTEXT_LOAD}}` | Shell command to load brain context at skill start |
+| `{{TOOL_REFERENCE}}` | Available mnemo commands table |
+| `{{PERSIST_BLOCK}}` | Shell commands for persisting knowledge at skill end |
+
+---
+
+## MCP Tools (58 total: 17 agent-facing + 42 specialized)
+
+Mnemo exposes **17 consolidated agent-facing tools** via MCP — designed to cover every workflow in minimal tool calls. Under the hood, these route to **42 specialized internal tools** for granular operations.
 
 ### Agent-Facing Tools (what the AI calls)
 
@@ -405,6 +441,8 @@ Mnemo exposes **16 consolidated agent-facing tools** via MCP — designed to cov
 | `mnemo_recall` | Load full project context at session start (budgeted ~2000 tokens) |
 | `mnemo_remember` | Store important context with auto-categorization & dedup |
 | `mnemo_decide` | Record permanent architectural decisions (never evicted) |
+| `mnemo_supersede` | Supersede (replace) an existing decision by ID |
+| `mnemo_redact` | Permanently redact a decision (e.g., accidental secrets) |
 | `mnemo_forget` | Delete a specific memory by ID |
 | `mnemo_search_memory` | Semantic search across memories (3-way RRF fusion) |
 | `mnemo_lookup` | 360° detail: class methods, function signatures, or full service architecture |
@@ -569,23 +607,26 @@ mnemo serve    # http://localhost:3333
 
 ```
 .mnemo/
-├── memory.json          Memories with retention scores & access history
-├── decisions.json       Permanent architectural decisions
-├── plans.json           Task tracking with dependencies
-├── context.json         Auto-detected project metadata
-├── graph.lbug/          LadybugDB knowledge graph (Kuzu engine)
-├── vectors_code.npy     ONNX embeddings of code symbols (384-dim)
-├── vectors_memory.npy   ONNX embeddings of memories
-├── meta_*.json          Vector metadata for cosine search
-├── engine-meta.json     File hashes for incremental detection
-├── parse-cache.json     AST parse cache (skip unchanged files)
-├── tree.md              Compact repo index (generated from graph)
-├── corrections.json     Wrong→right patterns with confidence
-├── lessons.json         Learned patterns with reinforcement
-└── slots.json           Pinned structured context (conventions, gotchas)
-```
-
-**Stack**: Python · LadybugDB (Kuzu) · ONNX Runtime · tree-sitter · Roslyn · NetworkX (Leiden)
+├── memory.json              Memories with retention scores & access history
+├── decisions.json           Permanent architectural decisions (computed snapshot)
+├── decisions.events.jsonl   Event-sourced decision log (source of truth)
+├── plans.json               Task tracking with dependencies
+├── context.json             Auto-detected project metadata
+├── learnings.json           Typed learnings with confidence & key-dedup
+├── autorun_state.json       SDLC orchestrator phase tracking
+├── graph.lbug/              LadybugDB knowledge graph (Kuzu engine)
+├── vectors_code.npy         ONNX embeddings of code symbols (384-dim)
+├── vectors_memory.npy       ONNX embeddings of memories
+├── meta_*.json              Vector metadata for cosine search
+├── engine-meta.json         File hashes for incremental detection
+├── parse-cache.json         AST parse cache (skip unchanged files)
+├── tree.md                  Compact repo index (generated from graph)
+├── corrections.json         Wrong→right patterns with confidence
+├── lessons.json             Learned patterns with reinforcement
+└── slots.json               Pinned structured context (conventions, gotchas)
+```
+
+**Stack**: Python · LadybugDB (Kuzu) · ONNX Runtime · tree-sitter · Roslyn · NetworkX (Louvain)
 
 **Zero cloud. Zero API keys. Zero telemetry. Everything runs locally.**
 
@@ -617,7 +658,7 @@ Pinned categories (never evicted): architecture, decision, preference
 ## CLI Reference
 
 ```bash
-mnemo init [--client CLIENT]     # Initialize in a repo (kiro, cursor, claude-code, amazonq, copilot)
+mnemo init --client CLIENT       # Initialize in a repo (kiro, cursor, claude-code, amazonq, copilot, generic)
 mnemo recall [--tier TIER]       # Show agent context (compact, standard, deep)
 mnemo map                        # Regenerate repo map from graph
 mnemo serve [-p PORT]            # Dashboard UI (default: 3333)
@@ -625,9 +666,33 @@ mnemo doctor                     # Diagnose installation issues
 mnemo reset                      # Remove Mnemo data (safe: only Mnemo-owned files)
 mnemo link [TARGET]              # Link another repo to multi-repo workspace
 mnemo remember "content" [-c CAT]# Store a memory
+mnemo learn -t TYPE -k KEY -i TXT# Store a typed learning (7 types, key-dedup)
+mnemo learnings [-t TYPE]        # List stored learnings (sorted by confidence)
+mnemo ingest [--file FILE]       # ETL: extract learnings from session transcripts
 mnemo tool NAME [--args]         # Call any MCP tool from CLI
 ```
 
+### Customizing What Gets Indexed
+
+Mnemo skips a built-in set of heavy/non-source directories during indexing
+(`node_modules`, `.venv`, `dist`, `build`, etc. — see `mnemo/config.py` for
+the full list).
+
+To skip additional directories in a specific repo, drop a `.mnemoignore`
+file at the repo root with one directory name per line:
+
+```
+# Heavy dirs to exclude from indexing
+data
+logs
+backups
+```
+
+Lines are matched by exact directory basename, anywhere in the tree (same
+semantics as the built-in list). Blank lines and `#` comments are ignored.
+Trailing slashes are tolerated. Glob / `.gitignore` semantics are not
+supported yet — patterns are basenames only.
+
 ---
 
 ## Contributing
@@ -636,7 +701,7 @@ mnemo tool NAME [--args]         # Call any MCP tool from CLI
 git clone https://github.com/Mnemo-mcp/Mnemo.git
 cd Mnemo
 pip install -e ".[dev]"
-pytest                    # 222 tests
+pytest                    # 641 tests
 ruff check .              # Lint
 mnemo init                # Test on self
 ```

{mnemo_dev-0.5.2 → mnemo_dev-0.6.0}/README.md

@@ -7,18 +7,18 @@
 <p align="center">
   <a href="https://pypi.org/project/mnemo-dev/"><img src="https://img.shields.io/pypi/v/mnemo-dev?style=flat-square&color=blue" alt="PyPI" /></a>
   <a href="https://www.npmjs.com/package/@mnemo-dev/mcp"><img src="https://img.shields.io/npm/v/@mnemo-dev/mcp?style=flat-square&color=red" alt="npm" /></a>
-  <a href="#"><img src="https://img.shields.io/badge/tests-222%20passing-brightgreen?style=flat-square" alt="Tests" /></a>
+  <a href="#"><img src="https://img.shields.io/badge/tests-641%20passing-brightgreen?style=flat-square" alt="Tests" /></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-AGPL--3.0-purple?style=flat-square" alt="License" /></a>
   <a href="https://pypi.org/project/mnemo-dev/"><img src="https://img.shields.io/pypi/pyversions/mnemo-dev?style=flat-square" alt="Python" /></a>
   <a href="https://marketplace.visualstudio.com/items?itemName=Nikhil1057.mnemo-vscode"><img src="https://img.shields.io/badge/VS%20Code-extension-007ACC?style=flat-square&logo=visualstudiocode" alt="VS Code" /></a>
 </p>
 
 <p align="center">
-  <code>[100% R@5]</code> <code>[2ms search]</code> <code>[58 tools]</code> <code>[16 agent-facing]</code> <code>[11 lifecycle hooks]</code> <code>[0 external DBs]</code> <code>[222 tests]</code>
+  <code>[100% R@5]</code> <code>[2ms search]</code> <code>[58 tools]</code> <code>[17 agent-facing]</code> <code>[11 lifecycle hooks]</code> <code>[0 external DBs]</code> <code>[641 tests]</code>
 </p>
 
 <p align="center">
-  <a href="#install">Install</a> • <a href="#why-mnemo">Why</a> • <a href="#benchmarks">Benchmarks</a> • <a href="#how-it-works">How It Works</a> • <a href="#supported-clients">Clients</a> • <a href="#features">Features</a> • <a href="#mcp-tools-16-agent-facing">Tools</a> • <a href="#dashboard-ui">Dashboard</a> • <a href="#architecture">Architecture</a>
+  <a href="#install">Install</a> • <a href="#why-mnemo">Why</a> • <a href="#benchmarks">Benchmarks</a> • <a href="#how-it-works">How It Works</a> • <a href="#supported-clients">Clients</a> • <a href="#features">Features</a> • <a href="#mcp-tools-17-agent-facing">Tools</a> • <a href="#dashboard-ui">Dashboard</a> • <a href="#architecture">Architecture</a>
 </p>
 
 ---
@@ -37,7 +37,7 @@ Session 2: you ask the agent to add a new endpoint. It already knows your servic
 ```bash
 pip install mnemo-dev    # or: brew tap Mnemo-mcp/tap && brew install mnemo
 cd your-project
-mnemo init              # defaults to Amazon Q (or: --client kiro, cursor, claude-code)
+mnemo init --client kiro # or: amazonq, cursor, claude-code, copilot, generic
 ```
 
 ---
@@ -89,8 +89,7 @@ Then initialize:
 
 ```bash
 cd your-project
-mnemo init                      # defaults to Amazon Q
-mnemo init --client kiro        # or: cursor, claude-code, copilot, generic
+mnemo init --client kiro        # or: amazonq, cursor, claude-code, copilot, generic
 ```
 
 **That's it.** Your agent now has persistent memory, semantic search, and architectural understanding.
@@ -215,13 +214,13 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 ## How It Works
 
 ```
-┌─── INIT (one-time, ~7s for 300 files) ──────────────────────────┐
+┌─── INIT (one-time, ~7s for 300 files) ───────────────────────────┐
 │                                                                  │
 │  1. Scan: single os.walk pass across repo                        │
 │  2. Parse: tree-sitter AST (14 langs) + Roslyn (C#)              │
 │  3. Graph: LadybugDB — files, classes, methods, CALLS edges      │
 │  4. Scope: cross-file function call resolution                   │
-│  5. Cluster: Leiden community detection                          │
+│  5. Cluster: Louvain community detection                         │
 │  6. Index: ONNX vector embeddings for semantic search            │
 │  7. Detect: languages, services, key classes, frameworks         │
 │  8. Configure: MCP server + hooks for your AI client             │
@@ -229,10 +228,10 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── SESSION START (automatic via hooks) ─────────────────────────┐
+┌─── SESSION START (automatic via hooks) ──────────────────────────┐
 │                                                                  │
 │  mnemo_recall injects into agent context:                        │
-│    • Architectural decisions (permanent, never evicted)           │
+│    • Architectural decisions (permanent, never evicted)          │
 │    • Hot memories (scored by access × recency × importance)      │
 │    • Active plan + next task                                     │
 │    • Compact repo index (classes per service)                    │
@@ -245,7 +244,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
                               ▼
 ┌─── DURING SESSION (tools + freshness) ──────────────────────────┐
 │                                                                  │
-│  Agent has 16 MCP tools available:                               │
+│  Agent has 17 MCP tools available:                               │
 │    • mnemo_lookup → full service/class architecture              │
 │    • mnemo_search → semantic search (code, memory, APIs)         │
 │    • mnemo_impact → blast radius if X changes                    │
@@ -258,7 +257,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── SESSION END (auto-capture via stop hook) ────────────────────┐
+┌─── SESSION END (auto-capture via stop hook) ─────────────────────┐
 │                                                                  │
 │  • Detects learnings (bug fixes, discoveries)                    │
 │  • Records session decisions                                     │
@@ -267,11 +266,11 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 └──────────────────────────────────────────────────────────────────┘
                               │
                               ▼
-┌─── BETWEEN SESSIONS (decay + maintenance) ──────────────────────┐
+┌─── BETWEEN SESSIONS (decay + maintenance) ───────────────────────┐
 │                                                                  │
 │  Every 10th recall:                                              │
 │    • Retention scored: salience × exp(-0.01 × days) + access     │
-│    • Hot (≥0.5) → Warm (≥0.25) → Cold → Evicted                 │
+│    • Hot (≥0.5) → Warm (≥0.25) → Cold → Evicted                  │
 │    • Contradictions auto-superseded (sim > 0.9)                  │
 │    • Low-value pruning (cap: 200 active memories)                │
 │    • Graph synced (stale memory nodes removed)                   │
@@ -279,7 +278,6 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 │  Pinned forever: architecture, decision, preference              │
 │                                                                  │
 └──────────────────────────────────────────────────────────────────┘
- └─────────────────────────────────────────────────────────────┘
 ```
 
 ---
@@ -288,6 +286,8 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 
 ### 🧠 Memory System
 - **Categorized storage**: architecture, pattern, bug, preference, decision, todo
+- **Event-sourced decisions**: JSONL append-only event log, computed snapshots, supersede/redact operations
+- **Branch-scoped decisions**: decisions tagged per branch, filtered on recall
 - **Retention scoring**: access frequency × recency × importance (Ebbinghaus-inspired decay)
 - **Branch-aware**: memories tagged with git branch, filtered on recall
 - **Contradiction detection**: new facts auto-supersede old conflicting ones (threshold: 0.6)
@@ -308,7 +308,7 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 - **Knowledge graph**: files, folders, classes, methods, functions, projects, communities
 - **14 languages**: Python, JS/TS, C#, Go, Java, Rust, Ruby, PHP, C/C++, Kotlin, Swift, Scala
 - **Roslyn enrichment**: C# method signatures, implements, full AST detail
-- **Leiden community detection**: automatic functional clustering
+- **Louvain community detection**: automatic functional clustering
 - **CALLS edges**: scope-resolved function call graph with confidence scoring
 - **Impact analysis**: upstream/downstream blast radius (N-hop BFS)
 - **Incremental freshness**: graph + vector index auto-update within 30s of file changes
@@ -342,9 +342,45 @@ Works with **any** agent that speaks MCP. One server, one memory, shared across
 
 ---
 
-## MCP Tools (58 total: 16 agent-facing + 42 specialized)
+## Workflow Skills & Orchestrator
 
-Mnemo exposes **16 consolidated agent-facing tools** via MCP — designed to cover every workflow in minimal tool calls. Under the hood, these route to **42 specialized internal tools** for granular operations.
+Mnemo includes **6 workflow skills** that guide agents through a structured SDLC pipeline, with **quality gates** enforcing standards between phases.
+
+### The Pipeline
+
+```
+investigate → plan → implement → verify → review → ship
+                                    ↑                ↑
+                              [tests_pass]    [tests_pass + plan_done + no_findings]
+```
+
+Each phase has a dedicated skill template with explicit instructions, shell commands for mnemo tools, and persistence hooks. The orchestrator tracks state in `.mnemo/autorun_state.json` and blocks advancement when gates fail.
+
+### Quality Gates
+
+| Gate | What it checks | Blocks |
+|------|---------------|--------|
+| `tests_pass` | Runs test suite (pytest/npm/maven/gradle) | review, ship |
+| `plan_done` | All plan tasks marked complete | ship |
+| `no_findings` | No unresolved critical findings in memory | ship |
+
+### Template Resolver System
+
+Skills use `{{RESOLVER}}` placeholders that get expanded per-host:
+
+| Resolver | Injects |
+|----------|---------|
+| `{{PREAMBLE}}` | Mandatory header — forces tool usage, DO NOT PROCEED pattern |
+| `{{LEARNINGS}}` | Past learnings relevant to current skill |
+| `{{CONTEXT_LOAD}}` | Shell command to load brain context at skill start |
+| `{{TOOL_REFERENCE}}` | Available mnemo commands table |
+| `{{PERSIST_BLOCK}}` | Shell commands for persisting knowledge at skill end |
+
+---
+
+## MCP Tools (58 total: 17 agent-facing + 42 specialized)
+
+Mnemo exposes **17 consolidated agent-facing tools** via MCP — designed to cover every workflow in minimal tool calls. Under the hood, these route to **42 specialized internal tools** for granular operations.
 
 ### Agent-Facing Tools (what the AI calls)
 
@@ -353,6 +389,8 @@ Mnemo exposes **16 consolidated agent-facing tools** via MCP — designed to cov
 | `mnemo_recall` | Load full project context at session start (budgeted ~2000 tokens) |
 | `mnemo_remember` | Store important context with auto-categorization & dedup |
 | `mnemo_decide` | Record permanent architectural decisions (never evicted) |
+| `mnemo_supersede` | Supersede (replace) an existing decision by ID |
+| `mnemo_redact` | Permanently redact a decision (e.g., accidental secrets) |
 | `mnemo_forget` | Delete a specific memory by ID |
 | `mnemo_search_memory` | Semantic search across memories (3-way RRF fusion) |
 | `mnemo_lookup` | 360° detail: class methods, function signatures, or full service architecture |
@@ -517,23 +555,26 @@ mnemo serve    # http://localhost:3333
 
 ```
 .mnemo/
-├── memory.json          Memories with retention scores & access history
-├── decisions.json       Permanent architectural decisions
-├── plans.json           Task tracking with dependencies
-├── context.json         Auto-detected project metadata
-├── graph.lbug/          LadybugDB knowledge graph (Kuzu engine)
-├── vectors_code.npy     ONNX embeddings of code symbols (384-dim)
-├── vectors_memory.npy   ONNX embeddings of memories
-├── meta_*.json          Vector metadata for cosine search
-├── engine-meta.json     File hashes for incremental detection
-├── parse-cache.json     AST parse cache (skip unchanged files)
-├── tree.md              Compact repo index (generated from graph)
-├── corrections.json     Wrong→right patterns with confidence
-├── lessons.json         Learned patterns with reinforcement
-└── slots.json           Pinned structured context (conventions, gotchas)
-```
-
-**Stack**: Python · LadybugDB (Kuzu) · ONNX Runtime · tree-sitter · Roslyn · NetworkX (Leiden)
+├── memory.json              Memories with retention scores & access history
+├── decisions.json           Permanent architectural decisions (computed snapshot)
+├── decisions.events.jsonl   Event-sourced decision log (source of truth)
+├── plans.json               Task tracking with dependencies
+├── context.json             Auto-detected project metadata
+├── learnings.json           Typed learnings with confidence & key-dedup
+├── autorun_state.json       SDLC orchestrator phase tracking
+├── graph.lbug/              LadybugDB knowledge graph (Kuzu engine)
+├── vectors_code.npy         ONNX embeddings of code symbols (384-dim)
+├── vectors_memory.npy       ONNX embeddings of memories
+├── meta_*.json              Vector metadata for cosine search
+├── engine-meta.json         File hashes for incremental detection
+├── parse-cache.json         AST parse cache (skip unchanged files)
+├── tree.md                  Compact repo index (generated from graph)
+├── corrections.json         Wrong→right patterns with confidence
+├── lessons.json             Learned patterns with reinforcement
+└── slots.json               Pinned structured context (conventions, gotchas)
+```
+
+**Stack**: Python · LadybugDB (Kuzu) · ONNX Runtime · tree-sitter · Roslyn · NetworkX (Louvain)
 
 **Zero cloud. Zero API keys. Zero telemetry. Everything runs locally.**
 
@@ -565,7 +606,7 @@ Pinned categories (never evicted): architecture, decision, preference
 ## CLI Reference
 
 ```bash
-mnemo init [--client CLIENT]     # Initialize in a repo (kiro, cursor, claude-code, amazonq, copilot)
+mnemo init --client CLIENT       # Initialize in a repo (kiro, cursor, claude-code, amazonq, copilot, generic)
 mnemo recall [--tier TIER]       # Show agent context (compact, standard, deep)
 mnemo map                        # Regenerate repo map from graph
 mnemo serve [-p PORT]            # Dashboard UI (default: 3333)
@@ -573,9 +614,33 @@ mnemo doctor                     # Diagnose installation issues
 mnemo reset                      # Remove Mnemo data (safe: only Mnemo-owned files)
 mnemo link [TARGET]              # Link another repo to multi-repo workspace
 mnemo remember "content" [-c CAT]# Store a memory
+mnemo learn -t TYPE -k KEY -i TXT# Store a typed learning (7 types, key-dedup)
+mnemo learnings [-t TYPE]        # List stored learnings (sorted by confidence)
+mnemo ingest [--file FILE]       # ETL: extract learnings from session transcripts
 mnemo tool NAME [--args]         # Call any MCP tool from CLI
 ```
 
+### Customizing What Gets Indexed
+
+Mnemo skips a built-in set of heavy/non-source directories during indexing
+(`node_modules`, `.venv`, `dist`, `build`, etc. — see `mnemo/config.py` for
+the full list).
+
+To skip additional directories in a specific repo, drop a `.mnemoignore`
+file at the repo root with one directory name per line:
+
+```
+# Heavy dirs to exclude from indexing
+data
+logs
+backups
+```
+
+Lines are matched by exact directory basename, anywhere in the tree (same
+semantics as the built-in list). Blank lines and `#` comments are ignored.
+Trailing slashes are tolerated. Glob / `.gitignore` semantics are not
+supported yet — patterns are basenames only.
+
 ---
 
 ## Contributing
@@ -584,7 +649,7 @@ mnemo tool NAME [--args]         # Call any MCP tool from CLI
 git clone https://github.com/Mnemo-mcp/Mnemo.git
 cd Mnemo
 pip install -e ".[dev]"
-pytest                    # 222 tests
+pytest                    # 641 tests
 ruff check .              # Lint
 mnemo init                # Test on self
 ```

mnemo_dev-0.6.0/mnemo/__init__.py

@@ -0,0 +1,5 @@
+"""Mnemo – Persistent memory and Code Intelligence For AI Agents."""
+
+from __future__ import annotations
+
+__version__ = "0.6.0"

{mnemo_dev-0.5.2 → mnemo_dev-0.6.0}/mnemo/api_discovery/__init__.py

@@ -131,3 +131,13 @@ def discover_apis(repo_root: Path) -> str:
     if len(lines) <= 2:
         return "No APIs discovered."
     return "\n".join(lines)
+
+
+def search_api(repo_root: Path, query: str) -> str:
+    """Search discovered APIs by query."""
+    full = discover_apis(repo_root)
+    if "No APIs" in full:
+        return f"No APIs matching '{query}'."
+    query_lower = query.lower()
+    lines = [ln for ln in full.split("\n") if query_lower in ln.lower() or ln.startswith("#")]
+    return "\n".join(lines) if any(not ln.startswith("#") for ln in lines) else f"No APIs matching '{query}'."

mnemo_dev-0.6.0/mnemo/breaking/__init__.py

@@ -0,0 +1,2 @@
+"""Backward-compat stub — real implementation in mnemo.quality.breaking."""
+from ..quality.breaking import *  # noqa: F401,F403