coordinator-mode
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill functions as an orchestration layer that decomposes user-provided tasks into instructions for sub-agents, creating an interpolation surface for untrusted data.
- Ingestion points: User requests enter the agent context and are processed in SKILL.md during the 'DECOMPOSE' and 'Worker Prompt Writing' phases.
- Boundary markers: The 'Worker Prompt Writing Guide' provides templates for sub-agent instructions but lacks explicit boundary markers or 'ignore embedded instructions' warnings to isolate user data from worker directives.
- Capability inventory: The skill is configured to use powerful tools including Bash, Write, Edit, and Agent (sub-agent spawning), which could be misused if malicious instructions are successfully passed to a worker.
- Sanitization: There are no instructions for sanitizing or escaping user input before it is inserted into the worker prompt templates, relying entirely on the coordinator agent's 'synthesis' to filter potential malicious content.
Audit Metadata