#1919th of 54 in Agent and MCP starters
nasiko
Control plane that tracks cost and routes LLM calls for coding agents
- Stars
- 12k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Feb 2026
- Language
- Rust
1,450 on 20 Apr 202611,964 on 9 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Nasiko is a single Rust server and CLI that discovers installed coding harnesses (Claude Code, Codex, OpenCode, Cursor CLI), reports their sessions and spend, and can route their LLM calls to a provider you configure. It also deploys A2A-speaking agents into Docker behind one proxy, with an MCP gateway, flow limits, ACLs and OTel traces. State lives in Postgres, Redis and S3-compatible storage.
Who it is for: Teams auditing and routing coding-agent spend and deploying A2A agents
Strengths
- Read-only `nasiko agents discover` changes no settings, needs no account, sends nothing off-machine
- Routes Anthropic-format Claude Code traffic to an OpenAI key, decoupling protocol from provider
- Agents get short-lived tokens, so no agent or log sees a real API key
- Cost and traces per agent and model via gen_ai OTel span attributes
Weaknesses
- Cursor CLI supports discovery and reporting only; LLM routing is not yet available
- Routing edits global per-user config, affecting every Claude Code or OpenCode process
- Needs Postgres, Redis and S3 backing services plus a running control plane
- OpenCode routing applies only to new sessions; existing sessions keep their model
What it needs
- Rust, OpenAI, Anthropic
- Needs Docker, Postgres, Redis, S3 (RustFS), OTel Collector (optional), Tempo (optional), Loki (optional), Rust toolchain (for CLI)
- env example file
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Quick answers
Is nasiko open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Can you run nasiko with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in Agent and MCP starters
See all 54| Rank | Project | Score |
|---|---|---|
| 1 | deepseek-harnessPlugin-based agent harness from DeepSeek with a local Web UI | 98 out of 100 |
| 2 | composioTypeScript and Python SDKs that give agents authenticated tools across 1000+ apps | 92 out of 100 |
| 3 | conductorDurable workflow engine that orchestrates microservices and LLM agents from JSON | 91 out of 100 |
| #4 | mcp-useTypeScript framework for building MCP servers with React Views and an Inspector | 87 out of 100 |
| #5 | a2uiProtocol and renderers for agents to stream native UI instead of text | 86 out of 100 |