The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill is designed to ingest and process user-provided or discovered legacy test assets to guide the creation and execution of new validation tests. This mechanism presents a surface where instructions embedded in external data could potentially influence agent behavior if not reviewed.
Ingestion points: User-provided test files or code snippets supplied in the request context as described in SKILL.md (§1.3.2).
Boundary markers: The instructions do not specify explicit delimiters or markers to isolate the logic of untrusted test assets from the agent's internal reasoning.
Capability inventory: The skill possesses the capability to execute shell commands via build tools (e.g., Maven, NPM), write new test specification files, and manage background application processes.
Sanitization: No specific filtering or validation logic is mentioned for the content of legacy test code before it is processed by the agent.
[System Command Execution]: To perform its primary validation and testing functions, the skill invokes standard development and infrastructure tools including Docker, Maven, Gradle, and Node.js.
Evidence: The skill documentation (SKILL.md Part 2) specifies using commands such as docker info, mvn spring-boot:run, npm start, and dotnet run for environment verification and application startup.
[External Resource Interaction]: The skill includes steps to verify network connectivity and download necessary testing dependencies from established technology service providers.
Evidence: SKILL.md (§1.3) describes connectivity checks to registry.npmjs.org and instructions for installing browser binaries using the Playwright CLI tool.

runtime-validation