The Agent Skills Directory

[SAFE]: The skill provides conceptual strategies (Anchored Iterative, Opaque, and Regenerative summarization) for managing long-running agent sessions without incorporating any harmful capabilities.
[SAFE]: The Python script scripts/compression_evaluator.py implements text processing using standard libraries (re, json) to extract metadata from conversation history. It does not perform network operations, access sensitive file paths, or use dynamic code execution functions like eval() or exec().
[SAFE]: The test suite tests/test_compression_evaluator.py uses importlib to load the local script for unit testing, which is a standard development practice for modular code validation.
[SAFE]: No obfuscation, prompt injection attempts, or persistence mechanisms were found in any of the provided files.

context-compression