self-review-diff
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local code changes, untracked files, and git diffs, creating a surface for potential instruction injection. * Ingestion points: git diff output, git status, and direct file reading as specified in SKILL.md. * Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in the analyzed data. * Capability inventory: The skill has the ability to write to the filesystem, perform git commits/rebases, and execute local test suites. * Sanitization: There is no mention of sanitizing or escaping the content of the changes being audited.
- [COMMAND_EXECUTION]: The skill uses shell commands to manage version control and run validation tools. * Evidence: Instructions include the use of git (status, branch, diff, ls-files, commit, rebase), the gh CLI (stack view, modify, rebase), and local project tools (lint, tests, type checks).
Audit Metadata