The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected across the skill's instructions or reference files. The skill functions as a logic-based reviewer for academic text.\n- [NO_CODE]: The skill consists entirely of Markdown-based instructions and rule files. It does not include any Python scripts, Node.js packages, or binary executables, eliminating risks associated with code execution.\n- [SAFE]: The skill incorporates robust internal constraints (Hard Rules) that explicitly prohibit the AI from fabricating references, PMIDs, DOIs, clinical evidence, or validation levels, which protects against hallucination-based misinformation in medical contexts.\n- [SAFE]: There are no network operations (curl, wget, etc.), external downloads from untrusted sources, or attempts to access sensitive system files (e.g., .ssh, .aws, .env).\n- [SAFE]: While the skill processes user-supplied manuscripts (Category 8 surface), it lacks any high-privilege capabilities—such as shell access, file system writes, or network requests—that would allow for exploitation through indirect prompt injection.

claim-strength-calibrator