The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run standard development commands such as pnpm openapi:lint and pnpm api:rebuild for validating documentation. These are routine development workflows and do not involve unsanitized input or dangerous execution patterns.
[DATA_EXFILTRATION]: While the skill contains code samples for API clients, they use safe placeholders like 'YOUR_API_TOKEN' and 'QUEUE_ID'. No sensitive credentials are hardcoded, and the skill does not facilitate sending local data to external servers.
[EXTERNAL_DOWNLOADS]: The skill does not perform any remote code downloads or external package installations. It relies on local project tools already present in the environment.
[PROMPT_INJECTION]: The skill uses natural instructional language to guide the agent. The use of 'CRITICAL' in the text refers to preventing manual errors in documentation files rather than attempting to bypass safety filters or override agent behavior.

api-doc