event-speaker-cold-outreach

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from sourcing packets, creating a surface for indirect prompt injection.\n
  • Ingestion points: The skill ingests a packet from samber/dev-event-organizer-skills@event-speaker-sourcing containing candidate names, evidence-backed "whys," and constraints (SKILL.md).\n
  • Boundary markers: The instructions mandate that the agent "Show the user the drafted invitation before it goes out" and confirm the completeness of the sourcing packet before proceeding, serving as human-in-the-loop safeguards (SKILL.md).\n
  • Capability inventory: The agent's capabilities are limited to drafting text for messages and recording interaction history if the harness supports persistent memory (SKILL.md).\n
  • Sanitization: The instructions do not define specific mechanisms for sanitizing or escaping the content of the sourcing packet fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 09:14 PM
Security Audit — agent-trust-hub — event-speaker-cold-outreach