The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill is configured to fetch protocol updates and specifications from external domains, including auth-md.com and official GitHub repositories associated with the WorkOS project. These downloads are used to ensure the agent has the most current version of the protocol documentation.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell-based commands using curl to interact with the isitagentready.com API for scanning domain readiness.
[INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection via its validation workflow.
Ingestion points: The skill accepts untrusted data by loading auth.md files from both local file system paths and remote URLs provided at runtime (Workflow: Validate in SKILL.md).
Boundary markers: The instructions do not explicitly mandate the use of delimiters or specific system instructions to ignore embedded commands within the ingested markdown files.
Capability inventory: The skill environment includes the ability to perform network requests and read local files to support validation logic.
Sanitization: The skill employs a comprehensive set of validation rules (Basic and Full levels) defined in references/validation-rules.md to check the structure, fields, and consistency of the provided data.

auth-md