skills/melumuccu/ai/gh-stack/Gen Agent Trust Hub

gh-stack

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the gh-stack extension via gh extension install github/gh-stack. This repository belongs to a well-known service provider and is the official source for the tool, which is considered a safe operation.
  • [INDIRECT_PROMPT_INJECTION]: The skill handles data from remote repositories, including branch names, commit messages, and pull request metadata, which could theoretically contain instructions intended for the agent.
  • Ingestion points: Remote PR metadata and branch names fetched during gh stack sync or gh stack view operations.
  • Boundary markers: No explicit delimiters are mentioned for parsing commit messages or PR descriptions.
  • Capability inventory: The skill utilizes shell execution for git and gh commands, including pushing to remote branches and modifying local Git configuration.
  • Sanitization: No specific filtering or sanitization of content from remote PRs or commit messages is described in the provided documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 07:34 AM
Security Audit — agent-trust-hub — gh-stack