summarize-meeting

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted meeting notes and transcripts, creating a vector where malicious instructions in the input could influence agent behavior.
  • Ingestion points: Source material such as raw notes, transcripts, or recordings described in SKILL.md.
  • Boundary markers: None explicitly suggested for the transcript input, though the skill provides 'Anti-Patterns' and 'Red Flags' documentation to guide the agent and user.
  • Capability inventory: The provided Python tools (action_item_tracker.py, decision_log_builder.py, and meeting_effectiveness_scorer.py) are restricted to reading JSON data and printing reports. They do not have capabilities to perform network requests, write to the file system, or execute system commands.
  • Sanitization: The skill relies on human review (Red Flag 9) and redaction passes (Red Flag 6) to ensure summary quality and security, but does not provide automated sanitization of the input.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 03:18 AM