develop
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill manages development environments using Git operations, specifically
git worktreefor parallel tasks andgit mergefor feature integration. - [COMMAND_EXECUTION]: Automated validation is performed by executing project-defined scripts through
npm testandnpm run lint. - [COMMAND_EXECUTION]: State tracking and lifecycle management are handled by running a local project automation script via
node .agents/scripts/orchestrator_state.js. - [PROMPT_INJECTION]: The skill processes user stories and product specifications (
artifacts/output/02-strategy/product-spec.md). This constitutes an indirect prompt injection surface where instructions embedded in specs could influence implementation; however, this is the core intended purpose of the skill. - Ingestion points:
artifacts/output/02-strategy/product-spec.md,requirements.md,user-stories.md. - Boundary markers: No explicit delimiters are specified to isolate document content from implementation instructions.
- Capability inventory: Subprocess execution (git, npm, node) and file system manipulation across project and worktree paths.
- Sanitization: The skill relies on the integrity of artifacts produced in earlier workflow phases.
Audit Metadata