recursive-context-pruning-token-budgeting
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for the agent to manage context and tokens efficiently. It contains no executable code, network operations, or credential access. It explicitly advises maintaining safety headers and original system instructions to prevent security issues during context pruning.
Audit Metadata