🦆🧠 Give your small robot a brain. Tell a Microduck, a Reachy Mini, a LeRobot arm or any ROS base what you want in plain language. An LLM (Claude, OpenAI, Gemini, Grok, or a local model via Ollama, vLL
Give your Microduck a brain. Any LLM, one .duck file. 🦆🧠 quackd, pronounced “quacked”. The brain daemon Microduck was missing, named like its siblings robotd, mediad, padd and tofd.
Tell a small robot what you want in plain language. An AI uses the robot's existing skills to do it.
"Find the ball and kick it", in the bundled simulator, driven by the scripted pilot (no API key). Same verbs, same safety layer, same perception as a real model run. See docs/assets.
quackd connects a small robot with two legs, the Microduck from Pollen Robotics, to a large language model (Claude, OpenAI, Gemini or Grok). The robot already knows how to walk, turn, kick, scoop something off the floor, look around and quack. quackd is the missing layer that turns a request like "find the ball and kick it" into the right sequence of those skills, watches what happens, and keeps going until the job is done or it is clearly impossible.
You do not need a robot to try it. A bundled simulator runs on any laptop in seconds. Goals that work today, in that simulator:
"Find the ball and kick it." · "Patrol, and quack twice if you see someone." · "Follow the person." · "Fetch the ball" (experimental, because the scoop is unreliable on purpose)
Goals like "find my keys" or "pick up the trash" are where this is going, not what it does yet. The robot ships at Christmas 2026 and nothing here has run on real hardware. The honest label for today is LLM driven, goal directed control of a simulated robot: an early, working step toward a small robot you can simply talk to.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/rokbenko--quackd.html)
Pulse — Hermes-style self-improving AI agent. Reliability-first rebuild with evaluated skill self-evolution, multi-agent team orchestration, dialectic user modeling, and fully self-hosted default stac
DrakTheDude/stackchan-local-llmYour StackChan, your language model, your network — local-first firmware and server kit for the M5Stack StackChan. Voice, face, camera and MCP tools, with nothing sent to a cloud service.
leluong141996-dev/Agent-Brain-HubShared, brain-inspired memory for AI agents: 13 brain regions with awake/sleep loops, a live visualizer, agents connected via REST/SDK/MCP, governance & audit, and business-value metrics. Works with C
reality-opened/openrealityOpen Reality: phone video to AI-queryable 3D scenes. MCP tools for Claude/Codex/Cursor (npm openreality-mcp), the self-hostable broker, and the VGGT-SLAM library, in one repo. BSD-2-Clause.
ismaelsoilet/jev-harnessZero-dependency System One decision harness: 5 semantic gates saving frontier AI agent tokens on trivial errors & doom loops. Python + TypeScript + Rust. MCP-compatible.
vinodborole/okf-kitTurn any website into a portable, agent-ready Open Knowledge Format (OKF) bundle — no LLM required. Crawl, sync, chat offline, and serve docs to coding agents over MCP.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.