subagent-driven-development
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes task descriptions and requirements from implementation plan files and interpolates them into the prompts for implementer and reviewer subagents.
- Ingestion Points: The
SKILL.mdfile reads plan files, and theimplementer-prompt.mdandspec-reviewer-prompt.mdtemplates receive the full text of these tasks. - Capability Inventory: Implementer subagents have the ability to modify project files, execute tests, and perform git operations (commit/add).
- Boundary Markers: While the prompts use Markdown headers to delimit task content, they lack explicit instructions for the subagent to ignore potential malicious directives embedded within the plan text.
- Sanitization: No evidence of validation or sanitization was found for the plan content before it is processed by the subagents.
Audit Metadata