Triple Memory

Complete memory system combining LanceDB auto-recall, Git-Notes structured memory, and file-based workspace search. Use when setting up comprehensive agent memory, when you need persistent context across sessions, or when managing decisions/preferences/tasks with multiple memory backends working together.

Security Scan

The skill's functionality matches a triple-backend memory system, but there are important inconsistencies (undeclared credentials used in examples) and privacy/scope concerns (automatic, silent capture and session-start syncing) that should be reviewed before installing.

Purpose & Capability

The skill claims to combine LanceDB, Git-Notes, and file-based search, which matches the instructions and included script. However the registry metadata declares no required credentials or env vars while the SKILL.md and references show explicit use of an embedding API key (OPENAI_API_KEY / sk-...) for LanceDB. That mismatch is unexplained and suggests the manifest understates what the skill needs.

Instruction Scope

Runtime instructions direct the agent to: read workspace memory files, run git-notes sync on every session start, auto-capture preferences/decisions, and operate 'silently' (do not announce memory operations). Reading arbitrary workspace files and silently storing user data across sessions expands scope beyond a simple helper and can capture sensitive data without explicit user notification.

Install Mechanism

No install spec (instruction-only) — low risk from arbitrary downloads. There is a small helper script (scripts/file-search.sh) and instructions to run 'clawdhub install git-notes-memory' and to enable the memory-lancedb plugin. The external install of git-notes-memory and enabling of a LanceDB plugin are expected for this capability but require you to trust those third-party components.

Credentials

The manifest lists no required env vars, yet SKILL.md and references show using OPENAI_API_KEY (and even an example 'sk-...' token) and rely on $WORKSPACE. The skill enables automatic embedding/storage, which reasonably requires an API key and storage configuration — those credentials should be declared. Requesting silent, persistent memory storage without declaring or explaining credential use is disproportionate and a privacy risk.

Persistence & Privilege

always:false and normal model invocation are used (good). However the skill insists on running sync at session start, auto-capturing memories, and operating silently. Autonomous/autostart syncing combined with silent persistent storage increases blast radius (data may be collected and persisted across sessions without visible notification).

What to consider before installing

Before installing, verify provenance and ask the author for clarifications: (1) Where exactly does LanceDB store embeddings and memories (local vs. remote)? Who can access them? (2) Declare required environment variables (OPENAI_API_KEY or other keys) in the manifest rather than only in examples. (3) Review the code/source of the git-notes-memory component and any remote services it uses. (4) Consider whether you want automatic, silent capture of workspace files and session context — if not, turn off autoCapture/autoRecall or disable silent operation. (5) Test in an isolated workspace (no secrets) first. If you need help, request the skill author add an explicit list of required env vars and a clear privacy/retention statement.

Triple Memory System

A comprehensive memory architecture combining three complementary systems for maximum context retention across sessions.

Architecture Overview

User Message
     ↓
[LanceDB auto-recall] → injects relevant conversation memories
     ↓
Agent responds (using all 3 systems)
     ↓
[LanceDB auto-capture] → stores preferences/decisions automatically
     ↓
[Git-Notes] → structured decisions with entity extraction
     ↓
[File updates] → persistent workspace docs

The Three Systems

1. LanceDB (Conversation Memory)

• Auto-recall: Relevant memories injected before each response
• Auto-capture: Preferences/decisions/facts stored automatically
• Tools: memory_recall, memory_store, memory_forget
• Triggers: "remember", "prefer", "my X is", "I like/hate/want"

2. Git-Notes Memory (Structured, Local)

• Branch-aware: Memories isolated per git branch
• Entity extraction: Auto-extracts topics, names, concepts
• Importance levels: critical, high, normal, low
• No external API calls

3. File Search (Workspace)

• Searches: MEMORY.md, memory/*.md, any workspace file
• Script: scripts/file-search.sh

Setup

Enable LanceDB Plugin

{
  "plugins": {
    "slots": { "memory": "memory-lancedb" },
    "entries": {
      "memory-lancedb": {
        "enabled": true,
        "config": {
          "embedding": { "apiKey": "${OPENAI_API_KEY}", "model": "text-embedding-3-small" },
          "autoRecall": true,
          "autoCapture": true
        }
      }
    }
  }
}

Install Git-Notes Memory

clawdhub install git-notes-memory

Create File Search Script

Copy scripts/file-search.sh to your workspace.

Usage

Session Start (Always)

python3 skills/git-notes-memory/memory.py -p $WORKSPACE sync --start

Store Important Decisions

python3 skills/git-notes-memory/memory.py -p $WORKSPACE remember \
  '{"decision": "Use PostgreSQL", "reason": "Team expertise"}' \
  -t architecture,database -i h

Search Workspace Files

./scripts/file-search.sh "database config" 5

Conversation Memory (Automatic)

LanceDB handles this automatically. Manual tools:

• memory_recall "query" - search conversation memory
• memory_store "text" - manually store something
• memory_forget - delete memories (GDPR)

Importance Levels

Flag	Level	When to Use
-i c	Critical	"always remember", explicit preferences
-i h	High	Decisions, corrections, preferences
-i n	Normal	General information
-i l	Low	Temporary notes

When to Use Each System

System	Use For
LanceDB	Conversation context, auto-retrieval
Git-Notes	Structured decisions, searchable by entity/tag
File Search	Workspace docs, daily logs, MEMORY.md

File Structure

workspace/
├── MEMORY.md              # Long-term curated memory
├── memory/
│   ├── active-context.md  # Current session state
│   └── YYYY-MM-DD.md      # Daily logs
├── scripts/
│   └── file-search.sh     # Workspace search
└── skills/
    └── git-notes-memory/  # Structured memory

Silent Operation

Never announce memory operations to users. Just do it:

• ❌ "I'll remember this"
• ❌ "Saving to memory"
• ✅ (silently store and continue)