Claw Mentor Mentee
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill is transparent about connecting to ClawMentor and making user-approved agent updates, but it has notable persistent and behavioral influence that users should review carefully.
Install only if you want ClawMentor to become a recurring source of agent configuration and behavior guidance. Review every proposed change before approving it, especially changes to skills, cron patterns, HEARTBEAT.md, SOUL.md, IDENTITY.md, or other core files. Protect the API key and remove the HEARTBEAT.md check if you no longer want automatic polling.
Findings (5)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Approved mentor updates could change how the agent is configured or what skills and scheduled behaviors it uses.
The skill is designed to bring externally supplied mentor packages into the local agent setup. That is the core purpose and is disclosed, but users should treat those packages as a supply-chain input.
Their package contains ... Annotated configuration, curated skills, cron patterns, security posture
Review each proposed mentor update before approving it, especially new skills, cron patterns, or security-related configuration changes.
External mentor advice could shape the agent's long-term behavior across future sessions if approved.
Mentor guidance can become persistent context that the agent consults later, including in core behavior files. The artifact says each item is approved by the user, making it purpose-aligned but still important to notice.
Proposes additions to `mentor-guidance.md` ... Identifies when mentor wisdom should go deeper — into SOUL.md, IDENTITY.md, HEARTBEAT.md, or other core files
Approve only guidance you want the agent to rely on persistently, and periodically review mentor-guidance and core behavior files.
After setup, the agent may continue contacting ClawMentor during heartbeat cycles until the heartbeat entry is removed or disabled.
The skill adds a persistent heartbeat instruction so the agent will keep checking ClawMentor automatically. This is disclosed and matches the update-notification purpose, but it creates ongoing behavior.
Schedule periodic checks — append this line to `HEARTBEAT.md` ... `- Check for new ClawMentor reports` ... run the mentee skill check on every heartbeat cycle
Only enable the heartbeat check if you want ongoing update polling, and remove the HEARTBEAT.md line if you later want to stop it.
Anyone with access to the stored key may be able to access the user's ClawMentor mentee account functions.
The skill asks for and stores a ClawMentor API key. This is expected for the service integration and is also declared as the primary credential.
When the user provides their API key: 1. Store it: `openclaw config set 'skills.entries["claw-mentor-mentee"].apiKey' THEIR_KEY`
Provide the key only if you trust the service and local agent environment, and rotate or revoke it if exposed.
ClawMentor can learn which mentor updates the user applied, skipped, or rolled back.
The skill discloses outbound reporting of user decisions to the provider. It also states raw configuration is not sent, so this is purpose-aligned but still a data flow users should understand.
Reports your decisions (applied/skipped/rolled back) back to ClawMentor — no config content shared
Be comfortable sharing update-decision telemetry with ClawMentor before using the service.