create-pr

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill guides the agent to execute a variety of shell commands for repository management and local verification. This includes git operations for branch management, the GitHub CLI (gh) for pull request creation, and various ecosystem-specific build and test tools (such as npm, pytest, cargo, and ruff) identified by scanning the project's configuration files.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests untrusted data from the repository's git history and file changes to summarize modifications for the pull request body.
  • Ingestion points: Step 1 uses git log and git diff to read change history from the current branch.
  • Boundary markers: None explicitly defined in the instructions, though the structured pull request template provides a natural context delimiter.
  • Capability inventory: The skill utilizes git for metadata retrieval, various local build, test, and lint tools discovered in the environment for code verification, and the gh CLI for remote repository interaction.
  • Sanitization: The skill does not specify sanitization for the content interpolated into the pull request body, relying on the agent's internal summarization logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 04:37 PM
Security Audit — agent-trust-hub — create-pr