skills/zlliang/skills/gh-cli/Gen Agent Trust Hub

gh-cli

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents the use of the gh command-line tool for GitHub operations. It correctly identifies destructive operations as high-risk and suggests verification steps prior to execution.- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it facilitates reading untrusted data from external sources.
  • Ingestion points: Processes external data from GitHub via gh issue view, gh pr view, gh run view, and gh pr diff (documented in SKILL.md).
  • Boundary markers: No explicit delimiters are suggested for data interpolation in the instructions, although the skill advises summarizing output.
  • Capability inventory: The agent can modify repositories, create/edit issues and pull requests, and execute arbitrary GitHub API calls through the gh tool.
  • Sanitization: The skill recommends summarizing potentially sensitive or untrusted output instead of verbatim reproduction to mitigate risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 03:52 PM
Security Audit — agent-trust-hub — gh-cli