The Agent Skills Directory

[SAFE]: The skill serves as a purely informational and utility resource for managing AI agent context, focusing on concepts like attention mechanics and progressive disclosure.
[SAFE]: No suspicious prompt injection patterns or attempts to override safety guidelines were found in the instructional content or metadata.
[SAFE]: The included script (scripts/context_manager.py) uses only standard Python libraries (typing, hashlib) for string manipulation and token estimation. It does not perform network operations, execute shell commands, or access sensitive system files.
[SAFE]: All file access is restricted to local directory paths for the purpose of loading summaries or detailed documentation as part of the progressive disclosure examples.
[SAFE]: There is no evidence of obfuscation, hidden instructions, or unverifiable external dependencies across any of the files.

context-fundamentals