wkc-setup
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the local repository during its discovery and archaeology phases. Malicious instructions embedded in source code or project documentation could potentially influence the generated specifications.
- Ingestion points: Step 1 reads existing project manifests (e.g., package.json, pyproject.toml) and documentation (docs/PRD.md). Step 6 performs a comprehensive survey of the codebase, including entry points and component logic, in SKILL.md.
- Boundary markers: The skill does not define specific boundary markers or instructions to disregard embedded commands when analyzing existing project files.
- Capability inventory: The skill can write configuration and documentation files, and it executes shell commands using git and the gh CLI for repository management and labeling.
- Sanitization: The instructions do not specify sanitization or filtering logic for content extracted from the codebase before it is used to generate the architecture spec.
- [COMMAND_EXECUTION]: The skill utilizes local CLI tools to manage the development environment.
- Evidence: The skill executes git init, git remote, git commit, gh auth status, gh repo create, and gh label create to configure the repository and issue tracker backend, as described in Step 2 and Step 5 of SKILL.md.
Audit Metadata