cmd-pr-description

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to determine the repository's base branch using gh repo view, git remote show, and git symbolic-ref.\n- [COMMAND_EXECUTION]: A potential command injection surface exists when the skill uses the $BASE_BRANCH variable in commands like git diff $BASE_BRANCH --stat. If the base branch name is fetched from a compromised repository configuration or provided by a user with malicious intent, it could contain shell metacharacters.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes untrusted content from the repository's history and diffs to generate a summary.\n
  • Ingestion points: Repository diffs and commit logs via git diff and git log (SKILL.md).\n
  • Boundary markers: Absent. The instructions do not provide delimiters or warnings to the agent to disregard instructions embedded within the git data.\n
  • Capability inventory: Execution of git and gh CLI tools (SKILL.md).\n
  • Sanitization: Absent. There is no evidence of filtering or escaping logic applied to the external git data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 02:40 AM
Security Audit — agent-trust-hub — cmd-pr-description