The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill functions as a template generator that interpolates user-supplied data (such as integration names, slugs, and API URLs) into various documentation and configuration files. This creates a surface for indirect prompt injection or schema confusion if malicious input is provided for these fields.
Ingestion points: User-provided integration name, slug, auth type, API base URL, and documentation URLs defined in the "Required Information" section of SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when interpolating these values into the file templates.
Capability inventory: The skill instructs the agent to perform multiple file-write operations for documentation (MDX), navigation (JSON), and provider configuration (YAML) files.
Sanitization: The skill does not define explicit sanitization or validation logic for the collected integration metadata before it is written to the file system.
[EXTERNAL_DOWNLOADS]: The generated documentation templates include instructions for installing the @nangohq/node package. This is an official resource provided by the vendor 'nangohq' for interacting with their API.

creating-integration-docs