meeting-action-items

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows security best practices by requiring human approval before the agent performs any external actions, such as creating tasks or sending messages.
  • [PROMPT_INJECTION]: The skill processes external meeting transcripts which are untrusted data sources. Ingestion points: Transcript files are ingested via read_file. Boundary markers: The instructions explicitly warn the agent to treat transcript content as data rather than instructions. Capability inventory: The agent can search, create, and update records in external platforms like Notion and GitHub. Sanitization: The skill requires all items to be traced to specific quotes or timestamps and includes a mandatory approval step for all external modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 07:40 AM
Security Audit — agent-trust-hub — meeting-action-items