The Agent Skills Directory

[SAFE]: The skill serves as documentation for the DeepRead API, communicating with official vendor domains (api.deepread.tech, www.deepread.tech).
[SAFE]: The authentication mechanism utilizes a standard Device Authorization Flow (RFC 8628) which is a secure method for CLI tools and agents to obtain credentials.
[SAFE]: Code examples provided for Python and Node.js use standard libraries and well-known, legitimate packages.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection (Category 8).
Ingestion points: Data and extracted text from processed documents are retrieved via GET /v1/jobs/{job_id} and GET /v1/preview/{token}.
Boundary markers: Absent; the API documentation does not specify the use of delimiters or 'ignore' instructions for the returned content.
Capability inventory: The skill is configured with allowed-tools: Bash, Read, Write.
Sanitization: Absent; extraction results are returned raw as processed by the backend models.

api