Agent Memory 1
v1.0.0Persistent memory system for AI agents to remember facts, learn from experience, and track entities across sessions with easy recall and updates.
⭐ 0· 83·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description, SKILL.md, CLI wrappers, tests, and src/memory.py all implement a local persistent memory (facts, lessons, entities) stored in an SQLite DB. There are no unrelated environment variables, binaries, or cloud credentials requested.
Instruction Scope
SKILL.md and examples only instruct creating/using the AgentMemory API and storing the DB at ~/.agent-memory/memory.db (or a custom path). There are no instructions to read unrelated system files, access secrets, call external endpoints, or exfiltrate data.
Install Mechanism
This is effectively an instruction/source bundle with no install spec that downloads arbitrary code. All source is included; there are no URL downloads, package installs, or extract steps in the provided files.
Credentials
The skill declares no required environment variables or credentials, and the code does not read environment secrets. It persists data to a user-local SQLite DB only.
Persistence & Privilege
always is false (not forced), model invocation is allowed (normal), and the skill only creates/uses a DB under the user's home directory. It does not alter other skills' configs or request elevated/system-wide privileges.
Assessment
This skill appears to be what it claims: a local SQLite-based memory for agents. Before installing, consider: 1) The DB is stored in ~/.agent-memory/memory.db by default and contains any facts you store — treat it as sensitive data (restrict filesystem permissions, do not put secrets there, or set a custom db_path). 2) SQLite FTS5 is used; ensure your Python/SQLite build supports FTS5 or the FTS parts may fail. 3) The code is included and readable — review src/memory.py if you have specific security/privacy requirements. 4) Run the bundled tests or run in an isolated environment first if you want to validate behavior. There are no signs of network exfiltration or unrelated credential requests.Like a lobster shell, security has layers — review code before you run it.
latestvk972a7446xhzk93nrwkxc8rqzd83dp7m
License
MIT-0
Free to use, modify, and redistribute. No attribution required.