ClawHub

Agentmm Skills

v1.0.1

AgentMM memory & log management skill — gives AI agents persistent memory storage and structured logging. Use when the user asks to remember information, rec...

⭐ 0· 122·1 current·1 all-time
by@cers-ai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description describe persistent memory and logging; required env var (AGENTMM_API_KEY) and required tools (curl, jq) directly match that purpose. The provided scripts implement read/write/search/logging operations against the declared API endpoints.
ℹ
Instruction Scope
SKILL.md and scripts restrict their actions to calling the AgentMM API and using a single small state file for the sync daemon (~/.agentmm_sync_state). They do not read arbitrary local files or other env vars. Note: the sync_daemon runs continuously (if started) and will repeatedly call /memory/changes and output results to stdout; by design this can cause ongoing transmission of remote data and requires operator awareness.
✓
Install Mechanism
No install spec — scripts are included and run directly. No downloads from external URLs or archive extraction occur. The SKILL.md suggests cloning a repo, but that is optional guidance and does not perform remote code execution at install time.
✓
Credentials
Only AGENTMM_API_KEY is required (plus optional AGENTMM_API_BASE). This is appropriate for an API-backed memory/logging skill. The scripts only read those environment variables and do not require unrelated credentials or secrets.
ℹ
Persistence & Privilege
always:false and no system-wide config changes are requested. The skill is intended to be invoked autonomously by the agent (default platform behavior). If the agent runs the sync_daemon, it will write a single timestamp file to the user home directory (~/.agentmm_sync_state). Consider enabling per-call approval if you want to prevent autonomous transmissions.
Assessment
This skill is coherent with its stated purpose: it will send any memories or logs you store to the AgentMM service (default https://api.agentmm.site) and requires AGENTMM_API_KEY in env. Before installing: (1) Confirm you trust the AgentMM service or self-host by setting AGENTMM_API_BASE to your instance; (2) keep AGENTMM_API_KEY secret and do not paste it into files; (3) be aware the included sync_daemon runs periodically and persists a single timestamp file (~/.agentmm_sync_state) — if you do not want autonomous network activity, do not start the daemon or enable require_approval in your OpenClaw configuration; (4) note a minor documentation inconsistency (the SKILL.md’s clone URL differs from the listed homepage) — verify the repository/source you clone if you fetch code manually.

Like a lobster shell, security has layers — review code before you run it.

latestvk9751kkzba9r45y5w63th99dhs83dhqv

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

🧠 Clawdis
EnvAGENTMM_API_KEY
Primary envAGENTMM_API_KEY

Comments