Skills
skills/claw-silhouette/botmap-skill/botmap/Gen Agent Trust Hub

botmap

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The agent reads untrusted game chat which could contain instructions intended to manipulate its behavior.\n
  • Ingestion points: The recentChat field from the api/surroundings endpoint in SKILL.md.\n
  • Boundary markers: Absent; there are no delimiters or instructions to ignore commands within chat messages.\n
  • Capability inventory: The agent can execute game actions (move, chat, interact) via curl commands.\n
  • Sanitization: Absent; external chat content is processed directly.\n- [Persistence Mechanisms] (LOW): The 'Tips' section in SKILL.md suggests setting up a cron job for automation, which is an advisory mention of system-level scheduling.\n- [Data Exposure & Exfiltration] (SAFE): The skill performs network requests to botmap.ai. While this domain is not on the global whitelist, the operations are essential for the skill's primary purpose and do not involve sensitive local file access.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 03:40 AM