The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several shell scripts (hooks/detect-playwright.sh, hooks/validate-test.sh) to automate project detection and test validation during the agent session.
detect-playwright.sh runs at session start to identify existing Playwright configurations.
validate-test.sh runs after file edits to check for anti-patterns (e.g., waitForTimeout, hardcoded URLs).
[DYNAMIC_EXECUTION]: The skill executes MCP servers for TestRail and BrowserStack using npx tsx on local source files. This is a standard pattern for MCP integrations but involves running TypeScript code directly from the plugin directory.
[INDIRECT_PROMPT_INJECTION]: The skill performs codebase analysis to understand the application structure before generating tests, creating a surface for indirect prompt injection if malicious instructions are embedded in the analyzed code.
Ingestion points: Uses Read, Grep, Glob, and LS tools (specifically in agents/test-architect.md and agents/migration-planner.md) to scan codebase files.
Boundary markers: No explicit instructions to ignore embedded prompts in project files were found in the agent definitions.
Capability inventory: Includes the ability to execute shell commands (via Bash permissions in settings.json), write files, and perform network requests through the MCP servers.
Sanitization: There is no explicit sanitization of codebase content before it is processed by the AI agents.

playwright-pro