rock-feedback

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes git and the gh (GitHub CLI) to manage branches, commits, and the creation of issues and pull requests. These operations are essential for the skill's purpose of automating feedback submission.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the xdlkc/rock-marketplace repository on GitHub. This destination is consistent with the author's identity and the stated goal of the skill. GitHub is a well-known service, and the interaction involves standard API/CLI usage.
  • [PROMPT_INJECTION]: The skill handles untrusted user input by processing feedback into structured GitHub reports.
  • Ingestion points: User descriptions of bugs or improvements (SKILL.md).
  • Boundary markers: The skill uses structured Markdown templates and explicitly recommends quoted shell heredocs (<<'EOF').
  • Capability inventory: Execution of git and gh shell commands.
  • Sanitization: Use of quoted heredocs prevents the shell from interpreting or expanding variables within the user-provided feedback, mitigating risks of indirect command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 05:51 PM
Security Audit — agent-trust-hub — rock-feedback