challenge-the-claim
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted content such as PR descriptions and user-provided claims.
- Ingestion points: The agent is instructed to 'inspect the description and changes relevant to the fixed claim' in SKILL.md.
- Boundary markers: The instructions explicitly direct the agent to 'separate their stated position from a charitable interpretation or your own implication' and to 'preserve the strength of that position'.
- Capability inventory: The skill is primarily instructional and does not perform direct shell execution; it references the use of another tool 'investigate-with-evidence' for factual checks.
- Sanitization: No explicit data sanitization or escaping mechanisms are defined.
- [NO_CODE]: The skill consists entirely of markdown instructions and YAML metadata. It does not include scripts, binaries, or automated installation commands.
Audit Metadata