The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the pnpm package manager to scaffold new testing infrastructure and run automated test suites.
[CREDENTIALS_UNSAFE]: The skill's instructions require the agent to analyze authentication storage mechanisms and manage test user credentials (e.g., seed-users.ts, JIRA_PAT). While high-privilege, this behavior is documented as necessary for the skill's primary function of auditing protected application routes.
[PROMPT_INJECTION]: The skill processes potentially untrusted data from external PRD files, creating a vulnerability to indirect prompt injection.
Ingestion points: The agent is instructed to read markdown files from the context-library/prds/ directory to define the audit scope and test logic.
Boundary markers: The skill does not provide specific delimiters or instructions to the agent to disregard natural language commands that might be embedded within the PRD content.
Capability inventory: The skill possesses the ability to modify the local filesystem by creating the apps/e2e/ package and to execute arbitrary shell commands via Playwright and pnpm.
Sanitization: There are no verification or filtering steps defined to sanitize the PRD content before it is used to generate executable test scripts or environment configurations.

e2e-audit