spec-driven-dev

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates a structured development process using project documentation files. It follows security best practices by requiring explicit user approval before implementation steps.
  • [COMMAND_EXECUTION]: Instructs the agent to use git add for documentation files, which is a standard and expected development operation.
  • [EXTERNAL_DOWNLOADS]: Provides links to the vendor's official website and GitHub repository for resources related to the skill's primary purpose. These are documented as vendor-owned resources.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads specification files from the local filesystem to guide its behavior.
  • Ingestion points: Markdown files within the docs/specs/ directory.
  • Boundary markers: The skill includes a clarification step (Step 3) and an ambiguity check (Step 2) to ensure requirements are clear.
  • Capability inventory: Reading documentation and proposing/implementing code changes.
  • Sanitization: The risk is mitigated by Step 6, which mandates waiting for explicit human approval of the implementation plan before any code implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 04:31 PM