The Agent Skills Directory

[Development Utility Execution]: The skill instructs the agent to utilize standard repository-specific tools such as make style, make quality, and python utils/generate_model_tests.py. These commands are used to ensure code standards and generate unit tests within the Diffusers project.
[External Material Processing]: The skill is designed to ingest reference repositories and inference scripts provided by the user. (1) Ingestion points: User-provided GitHub repository links and inference scripts (SKILL.md). (2) Boundary markers: The current instructions do not specify the use of explicit delimiters or warnings for the agent to ignore potentially malicious embedded instructions in the reference materials. (3) Capability inventory: The skill utilizes file-writing capabilities to create new model and pipeline files and shell execution to run test scripts (SKILL.md). (4) Sanitization: The process relies on the agent's ability to interpret and translate the logic from reference implementations into the framework's API without explicit automated sanitization steps.

integrating-models