The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes bun render-cards in the packages/docs directory. This is a local development command used to generate static assets for the project.
[INDIRECT_PROMPT_INJECTION]: The skill incorporates user-provided content into source code files, creating a potential surface for indirect prompt injection.
Ingestion points: The agent collects expert metadata (names, social handles, and descriptions) from the user and inserts them into packages/promo-pages/src/components/experts/experts-data.tsx.
Boundary markers: The instructions do not define explicit boundary markers or safety warnings for the agent to treat the input as untrusted data.
Capability inventory: The skill writes to the local file system and executes shell commands via the bun runtime.
Sanitization: No sanitization or validation steps for the expert's description or URLs are specified in the workflow.

add-expert