The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external sources, including TestRail API responses and user-supplied URLs or CSV references.
Ingestion points: Data enters the agent's context through qa-insights testrail commands that fetch mappings, suites, sections, and case details from TestRail. It also parses Plan and Run URLs provided by the user in the CollectBindings step (link-plans.md).
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential malicious content embedded in TestRail fields (e.g., case titles or summaries) before displaying them or using them in logic.
Capability inventory: The skill has the capability to write to the system (executing link commands) and read from the network (resolving URLs and fetching data via CLI tools).
Sanitization: There are no documented steps for sanitizing or validating the content of external data beyond ID and reference matching.
[COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to execute qa-insights and cawplan CLI utilities for data retrieval and execution.
Evidence: Commands like qa-insights testrail mappings get, qa-insights testrail link cases execute, and cawplan skill check are used throughout the workflows in SKILL.md and reference files.
[EXTERNAL_DOWNLOADS]: The skill facilitates interaction with external TestRail project instances to resolve artifact IDs and establish links.
Evidence: The skill parses and navigates to testrail_project_url, plan_url, and run_url as part of its core functionality for QA automation mapping.

cawplan-testrail-link