implement-spec
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill orchestrates a multi-step workflow by reading and acting upon external specifications and issue tickets. Since these inputs are untrusted, they provide a vector for indirect prompt injection where malicious instructions could attempt to control the subagents.
- Ingestion points: Reads spec and tickets to drive task graphs as documented in SKILL.md.
- Boundary markers: No delimiters or instructions to mitigate embedded command execution are defined for the processed documents.
- Capability inventory: Capabilities include file writing outside the repository (Step 2), worktree creation, git merges, and invocation of testing tools like tdd (Step 4).
- Sanitization: The workflow lacks sanitization or validation for external content ingested by subagents.
Audit Metadata