The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially influence agent behavior if the data contains malicious instructions.
Ingestion points: The skill reads repository pull request templates (.github/pull_request_template.md or similar) and processes user-supplied HTML files or dev-server URLs to generate screenshots.
Boundary markers: There are no instructions specifying the use of delimiters or warnings to ignore embedded instructions within the ingested content.
Capability inventory: The skill utilizes subprocess execution for the GitHub CLI (network and file operations) and Playwright (browser automation and screenshot generation).
Sanitization: The skill does not describe any validation, escaping, or filtering of the external content before it is processed or used in prompt construction.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of external tools and packages during setup.
Fetches the @aabuhijleh/gh-attach and @playwright/cli packages via the Bun package manager.
Installs additional agent skills from the microsoft/playwright-cli and cursor/plugins repositories.
Downloads browser binaries for Chromium using the Playwright CLI tool.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to automate the pull request process.
Uses gh pr comment to upload attachments and post comments to GitHub.
Uses playwright-cli screenshot and gh-attach shot to perform browser-based tasks and capture images of the application state.

writing-great-prs