ac-insight-extractor
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION] (SAFE): The script reads from local files such as '.claude/autonomous-log.jsonl' and writes to '.claude/insights.json'. There are no network calls or logic to transmit data externally.
- [REMOTE_CODE_EXECUTION] (SAFE): No remote code downloading, dynamic execution, or unsafe deserialization was found.
- [COMMAND_EXECUTION] (SAFE): The skill does not use subprocesses or system commands.
- [PROMPT_INJECTION] (SAFE): No malicious instructions aimed at overriding agent behavior were detected in the skill code or metadata.
Audit Metadata