hf-gradio

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly mentions a --token flag for accessing private Spaces, which encourages passing API tokens directly on the command line and thus would require the LLM to include secret values verbatim in generated commands/outputs.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to run "gradio info <space_id_or_url>" and "gradio predict <space_id_or_url>" against Hugging Face Spaces or public URLs, causing the agent to fetch and interpret JSON and other outputs from arbitrary public/user-generated Gradio apps (untrusted third-party content) which directly determine endpoints and payloads and thus can influence subsequent actions.