hunt-a-handle
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill includes instructions and example CLI commands for executing OSINT tools
sherlockandmaigret. (SKILL.md) - [PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of external data from social media profiles, creating a surface for indirect prompt injection. (SKILL.md)
- Ingestion points: User-provided handles and profile data fetched from hundreds of external platforms via
maigret. (SKILL.md) - Boundary markers: Directs users to define an authorized scope and review an ethics document before performing queries. (SKILL.md)
- Capability inventory: Uses shell commands to interact with external web services. (SKILL.md)
- Sanitization: Implements a rigorous verification process involving avatar matching, bio analysis, and a four-tier confidence grading system (Confirmed, Probable, Unconfirmed, Rejected) to evaluate findings. (SKILL.md)
- [NO_CODE]: The skill consists exclusively of markdown documentation and reference guides; it does not contain executable scripts, binaries, or automated code installation procedures.
Audit Metadata