krillclaw/KrillClaw

The world's smallest AI agent runtime. 49KB. Written in Zig. Zero dependencies.

41
/ 100
Emerging

KrillClaw is an AI agent runtime designed for small, resource-constrained devices like microcontrollers. It allows engineers to deploy autonomous AI agents that can interact with various LLMs, execute tools, and automate tasks directly on embedded hardware. It takes an instruction (like "create a REST API in Go") and produces the actions and code needed to achieve that goal, all on very low-power systems. This is for embedded systems engineers, IoT developers, and robotics engineers who need AI capabilities at the edge.

152 stars.

Use this if you need to run sophisticated AI agents on low-cost microcontrollers, IoT devices, or robotics platforms with minimal memory and processing power.

Not ideal if you are developing AI agents for desktop applications or cloud servers where resource constraints are not a primary concern and larger runtimes are acceptable.

embedded-systems IoT-device-management robotics-control edge-AI-deployment microcontroller-programming
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 11 / 25
Community 10 / 25

How are scores calculated?

Stars

152

Forks

10

Language

Zig

License

Last pushed

Mar 02, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/krillclaw/KrillClaw"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.