The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several shell commands, including git diff, grep, sed, find, and yq, to identify API changes and compare them against documentation. It also uses npx to execute the @apidevtools/swagger-cli validation tool. These are standard operations for a development-focused skill and do not involve remote script execution from unknown sources.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from the local repository which represents a potential injection surface.
Ingestion points: The skill reads contents from source files (e.g., **/routes/**/*.ts, **/api/**/*.py) and OpenAPI specifications (openapi.yaml) using grep, sed, and yq in SKILL.md (Step 3).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the output of the drift detection (MISSING variable) as untrusted data.
Capability inventory: The skill can execute shell commands, perform file system operations, and trigger other skills such as documentation-audit.
Sanitization: The skill does not perform sanitization or filtering of the strings extracted from the source code before presenting them to the agent as detected endpoints.

api-documentation