01
Context Compression for LLM APIs with TOMC
A runnable CPU-only example of task-oriented context preparation, with source inspection and clearly scoped token counting.
Learn · inspect · reproduce
Put TOMC to work.
Three focused guides for preparing context, resuming a project and measuring the result.
02
Reuse Project Notes Across Codex and Claude Code with TOMC
Save and recall explicit project notes across sessions and assistants using a shared local TOMC notebook.
03
How to Measure Context Compression and AI Agent Memory
Separate memory size, complete request input, answer quality and workflow cost when evaluating TOMC or another context preparation method.