meeting-action-items-extractor
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted meeting transcripts which may contain indirect prompt injections. Malicious instructions following markers like 'TODO:' are extracted verbatim into the JSON output. While the script only parses text, downstream agents using the extracted tasks could be influenced by this content. Ingestion points: File read via --input in meeting_action_items_extractor.py. Boundary markers: None. Capability inventory: Local file access only. Sanitization: None.
Audit Metadata