github

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script gh-skill.sh executes the official GitHub CLI tool to perform repository management tasks.
  • Evidence: Found in gh-skill.sh across all case branches (e.g., gh pr list, gh issue create, gh api).
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted data from GitHub repositories (Category 8).
  • Ingestion points: Content is ingested via pr-view, issue-view, and run-view commands in gh-skill.sh, which fetch PR descriptions, issue bodies, and CI logs.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat retrieved data as untrusted or to ignore embedded instructions.
  • Capability inventory: The skill has the capability to write to repositories (e.g., pr-create, pr-merge, issue-create), creating a risk if the agent is influenced by malicious data to perform unauthorized actions.
  • Sanitization: Data retrieved from GitHub is passed directly to the agent without filtering, escaping, or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 05:10 AM