Fact Checker
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input such as viral posts and quotes, which creates a potential surface for indirect prompt injection.
- Ingestion points: The agent is instructed to collect the 'exact claim' and pasted content from the user (SKILL.md).
- Boundary markers: The operating procedure explicitly instructs the agent to 'Isolate the claim' and 'Restate it as one or more precise, checkable propositions', which helps separate the user data from the agent's core instructions.
- Capability inventory: No executable scripts, subprocess calls, or network operations are defined within SKILL.md.
- Sanitization: The instructions focus on 'separating the factual core from interpretation or spin' to prevent the agent from processing instructional context embedded within the claim.
- [NO_CODE]: The skill is composed entirely of markdown instructions and YAML metadata; it does not distribute or reference any scripts, binaries, or automated tooling configurations.
Audit Metadata