arena
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting and synthesizing data produced by multiple sub-agents, creating a surface where malicious instructions in sub-agent output could influence the final synthesized artifact.\n
- Ingestion points: Sub-agent outputs produced in Phase B and stored in temporary directories or git worktrees are read and processed in Phases D and E (Graft phase).\n
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when the agent or sub-agents read content from the generated candidate files.\n
- Capability inventory: The skill utilizes the ability to spawn sub-agents with
run_in_background: true, read and write to the file system (specifically/tmpand git worktrees), and perform git operations.\n - Sanitization: No sanitization, escaping, or validation steps are mentioned for the content generated by sub-agents before it is integrated into the final output.
Audit Metadata