skills/microsoft/playwright-cli/dev/Gen Agent Trust Hub

dev

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution and External Downloads: The skill utilizes standard development tools including git, npm, and gh to manage the repository lifecycle. These operations involve downloading packages from official registries and interacting with Microsoft's official GitHub repositories. These are standard procedures for repository maintenance and development workflows.
  • Potential Indirect Prompt Injection Surface: The release preparation workflow involves fetching content (titles and descriptions) from external GitHub pull requests and issues using the GitHub CLI. This data is subsequently used to generate release notes. While this introduces a surface where external content could influence agent output, the risk is minimized as the resulting content is intended for inclusion in a pull request body that undergoes human review.
  • Ingestion points: Data is ingested from the output of gh pr view and gh issue view commands in release.md.
  • Boundary markers: No specific delimiters or instructions are used to separate the external data from the agent's instructions.
  • Capability inventory: The skill has capabilities to perform file writes, git push operations, and create pull requests via gh pr create.
  • Sanitization: The skill does not explicitly describe sanitization or filtering of the metadata fetched from GitHub before it is processed.
  • Dynamic Skill Management: The dependency update process includes a step to run a local script (scripts/update.js) which regenerates skill definitions from the codebase. This ensures the agent's operational context remains aligned with the latest version of the Playwright CLI tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:45 AM
Security Audit — agent-trust-hub — dev