Nasiko-Labs/nasiko

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

5330 stars · 1228 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Nasiko is a developer control plane for AI agents, built in Rust, that lets teams deploy, route, secure, and observe A2A-speaking agents with a single command. It combines a CLI for building and deploying agents with a dashboard for governance, access control, cost tracking (TokenOps), tracing, and multi-agent workflow orchestration.

Use cases

• track and forecast AI agent spend per agent and per model
• enforce access control and policies on what agents can touch
• route requests to the right agent across models and providers
• connect agents to MCP servers and tools with scoped permissions
• trace every agent interaction with cost attached
• orchestrate multi-agent workflows across frameworks
• deploy and version agents from the CLI
• monitor agent reliability, retries, and provider failovers

When to choose

• you run many production AI agents and need centralized governance, routing, and observability
• you need cost forecasting and budget enforcement for LLM usage across teams
• you want agent-to-agent calls proxied through a control plane for tracing and flow limits
• you need scoped MCP tool access and secrets injection per agent

When to avoid

• you only run a single simple agent or chatbot with no governance needs
• you want a lightweight local agent framework without a hosted control plane
• you need a fully self-contained open-source stack and cannot accept the non-standard license
• your stack does not speak A2A or MCP protocols

Facets

service · maturity active

agent-framework mcp api-gateway monitoring secrets-management authorization cli analytics tracing rate-limiting large-language-models developer-tools security monitoring self-hosted rust cli self-hosted cross-platform tokenops a2a-protocol agent-governance multi-agent-orchestration cost-tracking agent-registry control-plane llm-spend-management ai-agents docker web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Nasiko-Labs/nasiko")