TinyHarness 0.2.0 focuses on local AI coding and smaller context
TinyHarness is a local-first that uses Ollama by default. Code stays on the user's machine unless the user chooses another setup.
It tries to keep long AI agent sessions from becoming too large by combining tool results, keeping the small, and compacting old in stages. Its compaction feature is meant to shrink even a 1 million token conversation so it can be handled by a model with a 64,000 token .
Version 0.2.0 adds a with split panes for chat, live project structure, and a file tree, so the user can see what the agent is doing. It also adds Nix support, through an /image command, and installation through cargo install TinyHarness.
Key points
- TinyHarness runs as a local-first using Ollama by default.
- It includes features meant to stop long agent sessions from growing too large.
- Its compaction approach targets very long conversations that exceed a model's .
- Version 0.2.0 adds a split-pane with chat, project structure, and file tree views.
- The release also adds Nix support and through an /image command.