meeting-notes-formatter
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill ingests untrusted raw meeting notes which could contain malicious instructions aimed at influencing an AI agent. 1. Ingestion points: raw notes provided via string or external files. 2. Boundary markers: Not specified in the documentation. 3. Capability inventory: The skill can write files to the local system (formatter.save). 4. Sanitization: No input sanitization or filtering is mentioned in the documentation.
- EXTERNAL_DOWNLOADS (SAFE): The dependencies specified in requirements.txt (reportlab and python-dateutil) are established, reputable libraries in the Python community.
Audit Metadata