walkthrough
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a standard documentation assistant for developers. Its activities, including reading git logs and capturing screenshots, align with its stated purpose of creating walkthroughs.
- [COMMAND_EXECUTION]: The skill executes shell commands for repository management (git diff, git log, git status) and verification (go test, npx playwright). These operations are restricted to the local workspace and are appropriate for a development-focused agent.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the codebase, such as git diffs and test results. 1. Ingestion points: Reads repository metadata via git, test command outputs, and local documentation files. 2. Boundary markers: The skill uses clear Markdown headers and code blocks to separate ingested content in the final artifact. 3. Capability inventory: Uses Bash for git and testing tools; saves artifacts to the local tasks/ directory. 4. Sanitization: External data is wrapped in Markdown code blocks to prevent direct execution by the agent or subsequent processors.
Audit Metadata