The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied acceptance criteria from various document formats to generate test cases, which could be used to host malicious instructions.
Ingestion points: Reads content from specification documents including User Stories (US-XXX), Work Items (WI-XXX), and external files (Step 1).
Boundary markers: Not present. The skill does not employ specific delimiters or warnings to isolate the processed data from its own instructional logic.
Capability inventory: The agent has permissions to create and modify files within the repository and can interact with project management tools via Model Context Protocol (MCP) integrations.
Sanitization: No validation or escaping is applied to the acceptance criteria text before processing.
[COMMAND_EXECUTION]: The instructions require the agent to execute shell commands to manage its environment and document authoring.
Evidence: Instructs the agent to run a shell snippet (PLUGIN_ROOT="...") to resolve the absolute path of the plugin's root directory for reference loading.
Evidence: Uses git config user.name to retrieve user information for the "Creado por" field in generated documents.

test-define