clip-aware-embeddings
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides documentation and scripts for image-text matching with CLIP. No malicious code or unauthorized access patterns were found.
- [INDIRECT_PROMPT_INJECTION]: The validation script processes user input (queries) to recommend models. This is a standard functional surface with no identified exploit path.
- Ingestion points: The skill ingests untrusted data via command-line arguments in
scripts/validate_clip_usage.py. - Boundary markers: No explicit boundary markers or delimiters are used in the command-line argument processing.
- Capability inventory: The skill specifies access to Bash and file system tools via its frontmatter, though the included script only performs local text analysis.
- Sanitization: Input is processed using keyword matching and regular expressions to provide classification feedback and model recommendations.
Audit Metadata