agent-memory

Provides persistent memory for AI agents to remember facts, learn from experience, and track entities across sessions.

Security Scan

This package implements a local SQLite-based persistent memory for agents; its files, runtime instructions, and requested access are coherent with the stated purpose and do not request unrelated credentials or network access.

Purpose & Capability

Name/description (persistent memory, facts, lessons, entities) matches the provided code and SKILL.md. No unrelated binaries, cloud credentials, or external services are required.

Instruction Scope

SKILL.md and the CLI wrappers confine operations to storing/recalling/updating local memories and integrating with an agent lifecycle. There are no instructions to read unrelated system files, transmit data externally, or access secrets.

Install Mechanism

No install spec — code is bundled in the skill and has no external dependencies. That reduces remote code fetch risk. The package creates/uses a local SQLite DB as expected.

Credentials

The skill requests no environment variables or credentials. The only persistent resource it uses is a local SQLite DB (default ~/.agent-memory/memory.db), which is appropriate for a memory store.

Persistence & Privilege

The skill will create and write a database under the user's home directory by default and can be pointed at an arbitrary db_path. This is normal for a local memory store, but be aware a user-supplied db_path could point to sensitive locations (permissions permitting). It does not request always:true and uses normal autonomous invocation behavior.

Assessment

This skill appears to be what it says: a local SQLite-based memory for agents. Before installing, note: (1) it creates/writes a DB in ~/.agent-memory/memory.db by default — review and protect that file (file permissions, backups, encryption) if it will store sensitive info; (2) you can set db_path to any path — do not point it at system-critical files (e.g., /etc/passwd) or shared secrets; (3) the code does not perform network calls or request credentials, but any agent that uses this memory could export or transmit memories elsewhere — treat stored memories as data that your agents might surface; (4) there are minor code/quality issues (e.g., API surface hints in __init__ referencing get_memory) that are functional concerns but not indicators of malicious behavior. If you need stricter isolation, run the skill under a restricted user or in a container and review the DB contents regularly.

AgentMemory Skill

Persistent memory system for AI agents. Remember facts, learn from experience, and track entities across sessions.

Installation

clawdhub install agent-memory

Usage

from src.memory import AgentMemory

mem = AgentMemory()

# Remember facts
mem.remember("Important information", tags=["category"])

# Learn from experience
mem.learn(
    action="What was done",
    context="situation",
    outcome="positive",  # or "negative"
    insight="What was learned"
)

# Recall memories
facts = mem.recall("search query")
lessons = mem.get_lessons(context="topic")

# Track entities
mem.track_entity("Name", "person", {"role": "engineer"})

When to Use

• Starting a session: Load relevant context from memory
• After conversations: Store important facts
• After failures: Record lessons learned
• Meeting new people/projects: Track as entities

Integration with Clawdbot

Add to your AGENTS.md or HEARTBEAT.md:

## Memory Protocol

On session start:
1. Load recent lessons: `mem.get_lessons(limit=5)`
2. Check entity context for current task
3. Recall relevant facts

On session end:
1. Extract durable facts from conversation
2. Record any lessons learned
3. Update entity information

Database Location

Default: ~/.agent-memory/memory.db

Custom: AgentMemory(db_path="/path/to/memory.db")