gsp-subagent-build-loop
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines an orchestration workflow that processes user-supplied build tasks, creating a potential surface for instructions embedded in data to influence subagent behavior.
- Ingestion points: The workflow processes "meaningful build tasks" and "end-to-end slices" (SKILL.md) which are provided as external inputs to the agent.
- Boundary markers: The skill does not define specific delimiters or explicit instructions for subagents to ignore potential commands embedded within the build task content.
- Capability inventory: The process involves a "builder" subagent with implementation capabilities (file-write and code creation) and "reviewer"/"verifier" subagents with evaluative capabilities.
- Sanitization: There is no mention of sanitization or validation logic for the input task data before it is dispatched to the subagent loop.
Audit Metadata