TLDR: ultrasync replaces your coding agent's default search tool with one that does semantic search and returns the context it's looking for, including memory, with sub-second latency.
It does codebase understanding and utilizes network-speed regex instead of feeding everything to an LLM. Search and memory indexes can be built both lazily or eagerly. CLI for full visibility into your index. No repo pollution with Markdown file slop.
Please give it a try, honest feedback appreciated. Feel free to reach out here or on Discord with any questions.
TLDR: ultrasync replaces your coding agent's default search tool with one that does semantic search and returns the context it's looking for, including memory, with sub-second latency.
It does codebase understanding and utilizes network-speed regex instead of feeding everything to an LLM. Search and memory indexes can be built both lazily or eagerly. CLI for full visibility into your index. No repo pollution with Markdown file slop.
Please give it a try, honest feedback appreciated. Feel free to reach out here or on Discord with any questions.