essay-fact-checker
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Prompt Injection (SAFE): No instructions found that attempt to override system behavior or bypass safety filters. The instructions are strictly limited to fact-checking tasks.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials or access to sensitive file paths. The network activities (WebSearch, WebFetch) are aligned with the skill's primary purpose.
- Unverifiable Dependencies & Remote Code Execution (SAFE): The skill does not include any scripts, binaries, or package manager files. No remote code execution patterns were detected.
- Indirect Prompt Injection (SAFE): Ingestion points: 'claim_text' and 'context' in SKILL.md. Boundary markers: YAML structure is used for input/output. Capability inventory: WebSearch and WebFetch tools. Sanitization: None mentioned. Overall risk is low as the skill does not execute or evaluate the content being verified.
- Metadata Poisoning (SAFE): The YAML frontmatter and description are accurate and do not contain deceptive or malicious content.
Audit Metadata