The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill provides run-code and eval commands which allow the agent to execute arbitrary JavaScript strings within the browser context. This capability is used for advanced automation and inspecting DOM attributes.
Evidence: playwright-cli run-code "async page => { ... }" and playwright-cli eval "document.title" in SKILL.md and references/running-code.md.
[COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands such as npx playwright test to run local test suites and npm install for global tool installation. Executing local tests involves running code from the project directory.
Evidence: PLAYWRIGHT_HTML_OPEN=never npx playwright test and npm install -g @playwright/cli@latest in SKILL.md and references/playwright-tests.md.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to navigate to, snapshot, and search content from arbitrary external websites. This exposes the agent to malicious instructions embedded in web pages (e.g., in HTML comments or hidden text) that could attempt to influence its behavior or exfiltrate data retrieved via the tool's storage access capabilities.
Ingestion points: playwright-cli snapshot, playwright-cli find, and playwright-cli eval in SKILL.md.
Boundary markers: The skill instructs the agent to "redact sensitive values" and use "project-owned secure reference/path" for storage state in SKILL.md.
Capability inventory: run-code, eval, state-save, and shell command execution (npx playwright test) are available across all referenced scripts.
Sanitization: No automated sanitization of ingested web content is present; the skill relies on the agent's instructions to handle data safely.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of Playwright and its associated CLI tools from the official npm registry.
Evidence: npm install -g @playwright/cli@latest and npm init playwright@latest in SKILL.md and references/test-generation.md.

playwright-cli