The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using npm and npx to install dependencies and run the Playwright test suite. These are standard operations for its intended purpose of automated testing.
[DYNAMIC_EXECUTION]: The agent generates executable Playwright test scripts (.spec.ts files) at runtime based on the contents of replica/recon.md. These generated scripts are then executed using npx playwright test.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data to generate executable code.
Ingestion points: Reads flow definitions and data from replica/recon.md (Step 1).
Boundary markers: None present. The instructions do not define delimiters or provide warnings to ignore instructions embedded within the reconnaissance data.
Capability inventory: The agent has capabilities to write files to the e2e/ directory and execute shell commands (npm, npx).
Sanitization: There is no mention of sanitizing, escaping, or validating the input from recon.md before it is interpolated into the generated test code or the test plan.

replica-test