setup-github-labels

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to perform repository inspection and label management. Specifically, it uses gh repo view to identify the target, gh label list to assess the current state, and gh label create --force to apply the canonical set. These commands are integral to the skill's primary function and include explicit instructions for the agent to preview changes and obtain user confirmation before execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface by ingesting external data (existing label names and descriptions) from a GitHub repository via gh label list.
  • Ingestion points: Output of gh label list command (SKILL.md).
  • Boundary markers: The skill explicitly instructs the agent that "The tables below are the source of truth, not the GitHub UI," establishing a clear boundary between trusted local configuration and untrusted external state.
  • Capability inventory: File system write access (docs/agents/triage-labels.md, AGENTS.md) and GitHub API interaction (gh label create).
  • Sanitization: The process relies on the agent's reasoning to compare states and a mandatory human-in-the-loop confirmation step to validate the generated preview before any destructive or creative actions are taken.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:36 AM
Security Audit — agent-trust-hub — setup-github-labels