create-pr
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Command Execution: The skill makes extensive use of shell commands including
gitandgh(GitHub CLI). These are used to verify commit history, check for unstaged changes, and manage remote repository state, which is standard behavior for a PR management skill. - Repository API Interactions: The skill uses
gh apito query issue timelines and update pull request labels or reviewers. These interactions are scoped to thedocker/docsrepository on GitHub and represent the intended functionality for automating documentation contributions. - Dynamic Environment Discovery: It uses shell processing (via
sedandgit remote) to dynamically identify the fork owner. This allows the skill to work across different contributor environments without hardcoded user names. - Data Ingestion Surface: The skill reads external data such as git logs and GitHub issue timelines. While this introduces a surface for indirect prompt injection (e.g., from malicious commit messages), the skill uses structured processing with
jqand specific filters to minimize the risk of the agent misinterpreting data as instructions.
Audit Metadata