huggingface-tool-builder

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

• Third-party content exposure detected (medium risk: 0.30). In references/hf_model_card_frontmatter.sh, the required runtime workflow can ingest and parse outsider-authored free text from model README.md cards that hf download "$model_id" README.md fetches from Hugging Face, then parse_readme reads that downloaded content to extract YAML frontmatter.