discussion-composer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill functions as a writing aid for academic manuscripts. Analysis of the instructions and the included Python script reveals no evidence of malicious behavior, data exfiltration, or unauthorized system access.
- [SAFE]: The Python utility
scripts/main.pyis a simple CLI tool for generating discussion outlines. It uses only standard library components and lacks network or administrative capabilities. While it supports writing output to a file, the content is restricted to hardcoded template text. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted research data and questions (identified in SKILL.md Step 1 and scripts/main.py). Mandatory Evidence Chain: (1) Ingestion points: research findings and questions in SKILL.md and scripts/main.py; (2) Boundary markers: Absent; (3) Capability inventory: Local file-write in scripts/main.py, but no network or shell execution; (4) Sanitization: Absent. The risk is considered safe given the extremely limited capabilities available to the skill.
- [EXTERNAL_DOWNLOADS]: The documentation includes a reference link to the author's public GitHub repository at
github.com/aipoch/medical-research-skills. This is a standard reference for the skill's source code and does not involve automated downloads or remote code execution.
Audit Metadata