claim-strength-calibrator
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a source link to GitHub (github.com/aipoch/medical-research-skills). This reference to a well-known service is for informational purposes and does not involve any automated downloads or script execution.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-supplied data, specifically manuscript drafts and reviewer comments. Ingestion points: manuscript text, abstract, discussion, and reviewer comments processed as strings. Boundary markers: the skill defines a 'Scope Boundary' and 'Hard Rules' to constrain the agent's output, though no specific data delimiters are mentioned. Capability inventory: the skill is restricted to text generation and lacks network, file-system, or shell access. Sanitization: no explicit input sanitization is implemented, but the lack of executable capabilities makes this surface low-risk.
Audit Metadata