The Agent Skills Directory

[SAFE]: The skill provides conceptual documentation and diagnostic scripts for managing large context windows in AI agents. No malicious instructions or security bypass attempts were found.
[SAFE]: Analysis of the Python script degradation_detector.py reveals only diagnostic logic using standard libraries (numpy, re) without performing network operations, file system modifications, or command execution.
[SAFE]: The external references and author links point to the vendor's own official resources, and no typosquatting or impersonation patterns were identified.

context-degradation