Wayfinder
A Rust-based tool that optimizes LLM agent memory on complex projects by adding a semantic layer on top of an LLM Wiki. It improves agent knowledge over a codebase, cut token usage by 30% in the author's benchmarks, and includes features for handling concurrent agents updating shared memory to keep information reliable and fresh. Published to PyPI (installable via "uv tool install wayfinder-memory") and compatible with almost all agent harnesses.
What it addresses
Existing approaches to improving agent knowledge over a codebase weren't giving the results the author wanted, and they wanted to keep it simple while optimizing agent memory on complex projects.
Shared over time
Each entry reflects what was shared on that date. Details may have changed since then.
Wayfinder
A Rust-based tool that optimizes LLM agent memory on complex projects by adding a semantic layer on top of an LLM Wiki. It improves agent knowledge over a codebase, cut token usage by 30% in the author's benchmarks, and includes features for handling concurrent agents updating shared memory to keep information reliable and fresh. Published to PyPI (installable via "uv tool install wayfinder-memory") and compatible with almost all agent harnesses.
Problem: Existing approaches to improving agent knowledge over a codebase weren't giving the results the author wanted, and they wanted to keep it simple while optimizing agent memory on complex projects.
Posted by @zaterka · Original comment ↗
“I've been working on optimizing agent memory on complex projects. That resulted in this simple Rust-based tool called Wayfinder: https://github.com/zaterka/wayfinder”