The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard system utilities like grep, find, and Select-String to search through local OpenAPI specification files. These tools are used legitimately to locate endpoints and schema definitions within the user's workspace.
[INDIRECT_PROMPT_INJECTION]: The skill involves ingesting and interpreting data from external, potentially untrusted OpenAPI files, which presents a surface for indirect prompt injection. 1. Ingestion points: The workflow relies on reading spec.yaml and following recursive $ref definitions. 2. Boundary markers: The instructions do not provide explicit delimiters or safety guidance to ensure the agent ignores natural language instructions that might be embedded in schema description or summary fields. 3. Capability inventory: The agent is capable of executing file system searches and generating functional test code in YAML and Lua. 4. Sanitization: There is no mention of sanitizing or validating the contents of the OpenAPI spec before the agent uses it to generate test logic.

openapi-parser