skills/n8n-io/n8n/n8n:create-pr/Gen Agent Trust Hub

n8n:create-pr

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: Uses standard git and gh (GitHub CLI) commands to check repository status, push branches, and create pull requests. These operations are within the scope of the declared allowed-tools.
  • [DATA_EXPOSURE]: Explicitly instructs the agent to avoid disclosing security vulnerability details in public artifacts like branch names, PR titles, or commit messages, which is a defensive security practice.
  • [INDIRECT_PROMPT_INJECTION]: The skill interpolates content from the repository (branch names, plan files, and PR templates) into the pull request body.
  • Ingestion points: Reads files from .claude/plans/, .agents/plans/, and .github/pull_request_template.md, as well as branch names and git logs.
  • Boundary markers: Employs quoted heredocs ('EOF') in the command examples to prevent shell expansion of the generated body content.
  • Capability inventory: Can perform git push and gh pr create (write operations).
  • Sanitization: Relies on the AI to process and format templates based on gathered context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 06:55 AM
Security Audit — agent-trust-hub — n8n:create-pr