The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from an external source (TestRail) to generate local test code, which represents a potential attack surface if the source data is compromised.
Ingestion points: Fetches test case titles, preconditions, and steps via the testrail_get_cases tool (SKILL.md).
Boundary markers: The skill does not explicitly define sanitization or boundary markers when interpolating TestRail content into Playwright test templates.
Capability inventory: The skill writes generated test files to disk and executes them using npx playwright test (SKILL.md).
Sanitization: No specific sanitization logic is described for the imported text before code generation.
[COMMAND_EXECUTION]: The skill executes shell commands to perform testing operations.
Evidence: Invokes npx playwright test --reporter=json to generate results for synchronization (SKILL.md).
[SAFE]: Credentials and configuration are handled using secure practices.
Evidence: The skill requires TESTRAIL_URL, TESTRAIL_USER, and TESTRAIL_API_KEY to be provided as environment variables rather than hardcoding them, and it checks for their presence before proceeding.

testrail