implement-spec-agentteam
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external issue trackers which serves as an ingestion point for untrusted instructions.
- Ingestion points: Steps 1 and 4 involve reading the spec and tickets from an external issue tracker.
- Boundary markers: No explicit delimiters or boundary markers are defined to isolate untrusted ticket content in teammate prompts.
- Capability inventory: The skill spawns agents (via spawn_teammate) with the ability to modify the codebase, execute tests via the tdd tool, and merge branches.
- Sanitization: No content validation or sanitization is performed on the ingested ticket data before it is passed to implementer teammates.
Audit Metadata