What problem it solves

Coding agents often run with separate dashboards, billing units, and provider defaults, making usage difficult to attribute to an agent, model, or outcome. Teams also need to apply routing, permissions, cost controls, and observability without replacing the harness workflows developers already use.

Key features

• Non-invasive agent discovery — The CLI detects supported coding-agent harnesses, reports their versions and configuration locations, and can do so without changing settings or sending data off the machine.

  Source: README

• Session reporting and cost attribution — Optional hooks record coding-agent activity, queue completed turns locally, and deliver them to the control plane. Token and cost information can then be attributed per agent and model.

  Source: README

• Decoupled LLM routing — Developers can route an agent's inbound wire protocol to a selected provider and model, while agents receive short-lived identity tokens rather than direct provider keys.

  Source: README

• Policy-enforcing call path — A single ingress proxies inter-agent calls so access controls, flow guards, and tracing can be applied at each hop. The platform also provides a permission-filtered MCP Gateway for tools.

  Source: README

• Agent deployment and orchestration — The platform can deploy agents that speak A2A through an embedded registry, while its routing pipeline selects agents using semantic shortlisting, contextual reranking, and a final LLM selection.

  Source: README

Quick start

Copied verbatim from the project's README.

git clone https://github.com/Nasiko-Labs/nasiko.git && cd nasiko
cargo install --path cli/ --force
nasiko agents discover

git clone https://github.com/Nasiko-Labs/nasiko.git
cd nasiko
cp .env.example .env

docker compose up -d

Who it may suit

This project may suit teams operating multiple coding-agent harnesses and needing centralized usage attribution, model routing, policy enforcement, and agent deployment. It may also suit developers who want to retain their existing harness commands and interfaces.

Topics to discuss

• Which coding-agent reporting events should be collected by default when teams need attribution without recording prompt or response content?

• How should provider routing policies be defined for teams that use different models across Claude Code, Codex, and OpenCode?

• Which agent deployment and governance capabilities should the community prioritize next?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/Nasiko-Labs/nasiko#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.