prompt-compression
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists of documentation and metadata files. A review of the instructions and configuration reveals no malicious patterns, such as prompt injections, unauthorized data access, or obfuscated code.
- [NO_CODE]: The skill does not bundle any executable scripts, binaries, or automated workflows. It serves as an architectural guide for utilizing external libraries like llmlingua and transformers to optimize prompt lengths.
Audit Metadata