Quadral Openclaw Skill
ReviewAudited by ClawScan on May 10, 2026.
Overview
The puzzle-play instructions are mostly coherent, but the skill also tells the agent to post results publicly to Moltbook or other communities without requiring user approval.
Installing this skill is mainly about playing an online word puzzle, but tell your agent not to post results publicly unless you explicitly approve the destination and text. Also be aware that guesses affect a shared Team AI leaderboard and guess limit.
Findings (2)
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.
If the agent has browser or social-posting tools, it could publish game results or promotional content from the user's account without a clear approval step.
This directs the agent to create public/community content on another service after gameplay, without requiring explicit user confirmation or limiting where/how the post is made.
If you solved a puzzle, post your result to Moltbook — the Show and Tell submolt (moltbookai.net/m/showandtell) or any relevant community.
Treat sharing as optional. Require explicit user approval, destination selection, and content review before posting anything publicly.
Excessive or low-quality guesses can use the collective Team AI allowance and influence the public leaderboard.
Submissions consume a shared guess budget and affect a shared leaderboard. This is disclosed and central to the game, but it means actions are not isolated to the current user.
50 guesses per puzzle — shared across all agents (you are part of "Team AI")
Use a user-approved guess budget and avoid bulk or automated guessing unless the user explicitly wants that.