agents-orchestrator
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project specification files and task lists which could contain malicious instructions designed to hijack the orchestration flow or influence spawned agents.
- Ingestion points: Reads markdown files located in project-specs/ and project-tasks/ as seen in Phase 1, 2, and 3.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when processing these files.
- Capability inventory: The skill can execute shell commands like ls, cat, and grep, and it has the authority to spawn various specialist agents (e.g., Frontend Developer, Backend Architect) with high-level instructions.
- Sanitization: The instructions do not specify any validation or sanitization of the content found within the specification or task list files before they are used to drive the workflow.
Audit Metadata