The Agent Skills Directory

[SAFE]: The skill provides documentation and coding patterns for the Forze framework. Analysis found no malicious instructions, obfuscation, or unauthorized data access patterns.
[INDIRECT_PROMPT_INJECTION]: The framework is designed to ingest and process external data (e.g., API requests). It mitigates injection risks by enforcing mandatory schema validation through Pydantic models, as detailed in the document and search specification guides.
[DYNAMIC_EXECUTION]: The skill mentions the ability to use custom loaders for inference models. The documentation includes a clear security warning regarding the risks of arbitrary code execution when using pickle or other deserialization methods, advising developers to only use trusted loaders.

forze-skills