review-branch

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git and gh (GitHub CLI) commands to identify the default branch, compute merge bases, and generate diffs of staged and unstaged changes.\n- [EXTERNAL_DOWNLOADS]: Fetches commit metadata from GitHub's official API using gh api to verify that the local branch matches the remote state. This is a neutral operation targeting a well-known service.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it processes repository-controlled data from files and diffs during the review process.\n
  • Ingestion points: Reads repository documentation (AGENTS.md, CLAUDE.md) and the entire content of the code diff for review.\n
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the analyzed repository files.\n
  • Capability inventory: The skill uses git and gh for read-only repository inspection; findings are reported back to the user via the agent.\n
  • Sanitization: There is no evidence of sanitization or filtering of the file content before it is processed by the reviewer model.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 02:08 PM
Security Audit — agent-trust-hub — review-branch