orchestrate

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The analyzed skill contains only markdown-based instructions and metadata. It does not include any scripts, binaries, or automated configuration files.
  • [PROMPT_INJECTION]: The skill outlines a workflow that creates a surface for indirect prompt injection by having agents research a codebase and perform actions based on those findings.
  • Ingestion points: Data enters the agent context during the 'Research' phase (SKILL.md) when the scout or orchestrator agents explore the codebase.
  • Boundary markers: There are no instructions or delimiters provided to help the agent distinguish between code logic and potential embedded instructions in the files being researched.
  • Capability inventory: The 'orchestrator' agent is granted 'Full tool access', and the 'reviewer' agent has 'bash' access (SKILL.md).
  • Sanitization: No sanitization or verification steps are defined to filter or validate content retrieved from the codebase before it influences planning or implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 12:48 AM