Oko: local MCP search tool cuts coding agents' token use by up to 46%
TheMoonMidas · x · 2026-09-20
- Oko (open source, Rust) is a local MCP search tool that helps Codex, Claude Code, and OpenCode find relevant code faster and read less irrelevant code.
- Early testing shows up to 46% lower token usage; gains vary by task and tool.
- It ranks source snippets using TypeSafe AI's Jev (API key required); a local keyword-search mode works without a key.
- Usable via coding agents or CLI (e.g. oko ask "where is authentication handled?"), with prebuilt binaries requiring no Node/Rust runtime.
More from coding & agent
- Devs debate running stateful AI agent runtimes on Cloudflare Workers and other edge runtimes — merlinofthewater · 2026-09-20
- Jev's instant compression scores each tool call to trim agent context without summarization — FinanceYF5 · 2026-09-20
- Stop using LLMs for ticket triage: Jev turns unstructured input into an executable smart if — sven_ai · 2026-09-20
- Jeff Dean's 1-Hour AI Engineering Lecture: From LLM Basics to Agent Graphs — irinarish · 2026-09-20
- Multi-agent systems work best with clear roles, not more agents — _jaydeepkarale · 2026-09-20
- Jev founder: all AI models are built for human-in-the-loop, not true automation — hardimanjames · 2026-09-20