board-minutes

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its document processing and summarization workflow.
  • Ingestion points: Meeting agendas, slide decks, and pre-read materials provided by the user (Step 3) are ingested and summarized to populate the minutes draft.
  • Boundary markers: The instructions do not define specific delimiters, XML tags, or "ignore embedded instructions" warnings for the content extracted from external materials.
  • Capability inventory: The skill performs file writes to local matter-specific folders (~/.claude/plugins/config/claude-for-legal/corporate-legal/matters/) and accesses meeting metadata via calendar connectors.
  • Sanitization: There is no explicit requirement to sanitize or validate the content of user-uploaded files before they are processed by the agent, allowing malicious instructions embedded in documents to potentially influence agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 03:58 PM
Security Audit — agent-trust-hub — board-minutes