w

waku-agent

Waku Agent is a locally-first AI assistant with looping, memory, and evaluation capabilities.

🌍 OverseasFreeOpen sourceGeneral开源桌面操作
Platforms: DesktopWeb
Region
Overseas
Pricing
Free
Open source
Yes
GitHub Stars
★ 1.1k
Source
GitHub
Added
2026-08-11
Last verified
2026-08-11
waku-agent

Overview

Waku Agent is a locally-first personal AI assistant designed to give users full ownership and control. It demonstrates the four foundational pillars behind every serious agent: Harness, Loop, Memory, and Eval/LLM-Ops. The codebase is concise and readable, allowing users to run it on their own laptops and interact with it via simple commands. Core features include locally stored memory files, a short 95-line Python loop, built-in evaluation mechanisms, and a local dashboard that provides real-time visibility into message flows and system status.

Key features

  • Locally-first approach with complete user control over data
  • Concise 95-line Python loop code
  • Built-in evaluation mechanism supporting multiple models
  • Local dashboard for real-time system state visualization

Use cases

Personal task managementSchedulingInformation retrievalAutomated tool invocation

Pros

  • Simple and readable code
  • Strong data privacy protection
  • Supports multiple language models
  • Real-time system state visualization

Limitations / notes

  • Requires some programming knowledge
  • Certain features require internet connectivity

Who it's for

DevelopersTech enthusiastsIndividual users

This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.

FAQ

Is it free?

Yes, Waku Agent is open-source and free.

Is Chinese support available?

Documentation and community support are available in Chinese, but specific functionality may require an English environment.

Can it be used commercially?

Yes, but subject to compliance with the open-source license agreement.

Similar agents

Something wrong? Let us know on the About page and we'll fix it.