skills/gannonh/skills/orchestrate/Gen Agent Trust Hub

orchestrate

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Analysis of the skill instructions found no evidence of prompt injection, data exfiltration, obfuscation, or malicious command execution.
  • [SAFE]: The skill includes explicit security best practices, such as instructing the agent not to send secrets, credentials, or private data unless authorized.
  • [SAFE]: (Indirect Prompt Injection Surface): The skill coordinates multiple agents and integrates their outputs, which is a potential surface for indirect injection. However, it incorporates robust safety practices:
  • Ingestion points: Processes worker results and repository content (SKILL.md).
  • Boundary markers: Employs goal-style contracts with explicit 'In-scope' and 'Out-of-scope' boundaries to limit sub-agent behavior.
  • Capability inventory: Involved in code merging, test execution, and cloud agent dispatch.
  • Sanitization: Recommends intentional integration, manual inspection of diffs, and verification via linting, typechecking, and automated builds before finalizing work.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 03:08 PM
Security Audit — agent-trust-hub — orchestrate