renovate-review

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from untrusted GitHub Pull Request data. \n
  • Ingestion points: The skill retrieves PR titles and bodies using gh pr view as specified in SKILL.md. \n
  • Boundary markers: There are no delimiters or instructions to the agent to ignore potentially malicious commands embedded in the PR data. \n
  • Capability inventory: The availability of Bash, WebFetch, and Read tools creates a surface for exploitation if injected instructions are followed. \n
  • Sanitization: Content retrieved from the PR is processed without validation or sanitization. \n- [COMMAND_EXECUTION]: Shell commands utilize direct variable interpolation, which presents a command injection risk. \n
  • Evidence: The instruction gh pr view $ARGUMENTS in SKILL.md passes raw user-supplied arguments to the shell. \n
  • Impact: Malicious input containing shell metacharacters could lead to the execution of unintended commands. \n- [EXTERNAL_DOWNLOADS]: The skill fetches release information from GitHub, which is a well-known service provider. \n
  • Evidence: It uses WebFetch to access release pages at https://github.com/<owner>/<repo>/releases. \n
  • Context: This activity is a legitimate part of the skill's functionality to assess dependency changes.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 05:39 AM
Security Audit — agent-trust-hub — renovate-review