subagent-driven-development
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were detected. The skill uses subagent templates to manage complex development tasks in a structured manner. The templates include proactive safety instructions, such as the requirement for reviewers to verify code independently rather than trusting implementer reports.\n- [INDIRECT_PROMPT_INJECTION]: The skill handles task descriptions from external implementation plans, presenting a standard ingestion surface necessary for development workflows.\n
- Ingestion points: Tasks are extracted from implementation plans (e.g.,
docs/plans/feature-plan.md) inSKILL.md.\n - Boundary markers: Markdown headers are used in
implementer-prompt.mdandspec-reviewer-prompt.mdto delimit task content, though no specific instructions are provided to ignore instructions embedded within the tasks.\n - Capability inventory: Subagents can implement code, run tests, and perform reviews using the
general-purposeandsuperpowers:code-reviewertools.\n - Sanitization: Task text is passed to subagents for execution as defined by the workflow without additional sanitization.
Audit Metadata