AB
AiBoss
project

OpenClacky - A low-cost AI agent open-sourced by Li Yafei's team

OpenClacky is an open-source AI agent developed by Li Yafei's team, supporting one-click installation and cross-platform operation. The tool significantly reduces usage costs through a streamlined toolset, intelligent context compression, and model routing, allowing professionals to run it around the clock...

What is OpenClacky?

OpenClacky is an open-source AI agent developed by Li Yafei's team, supporting one-click installation and cross-platform operation. The tool significantly reduces usage costs through a streamlined toolset, intelligent context compression, and model routing, allowing professionals to work around the clock with an AI assistant. OpenClacky integrates with a built-in browser and instant messaging, supports skill packaging and sales, and aims to enable every industry expert to become a micro-SaaS founder with zero coding.

Main functions of OpenClacky

  • Intelligent task executionIt utilizes 16 core tools to accomplish various tasks such as programming, data analysis, and content creation.
  • Multimodal generationSupports the creation and editing of all modal content, including text, images, and videos.
  • Browser automationThe built-in browser automatically captures web page data, searches for information, and organizes and outputs it.
  • IM tool integrationSeamlessly integrates with communication tools such as Lark, WeChat, DingTalk, and Discord to automatically push results.
  • Skill MarketIt supports packaging industry experience into Skills for distribution, updating, and monetization.
  • Model switchingSupports platforms such as BYOK and Alibaba Cloud Bailian, enabling one-click access to different modal model capabilities.
  • Scheduled task hostingIt can run continuously in the background and automatically perform daily tasks such as scheduled searches and data scraping.

OpenClacky's technical principles

  • Simplified toolsetOnly 16 core tools are retained, and long-tail capabilities are handed over to Skills for on-demand invocation, avoiding the use of context tokens by schemas of unused tools.
  • Intelligent Context CompressionDuring idle periods, the historical dialogue context is automatically compressed to prevent long dialogues from being repeatedly passed in unchanged, thus reducing continuous operating costs.
  • BYOK Model RoutingThe model is dynamically selected based on the complexity of the task. For complex tasks, a strong model is used, while for simple tasks, a cheaper model is used, thus achieving refined cost control.
  • Self-evolution Skill AccumulationThrough two self-evolving hooks, high-frequency operations are automatically accumulated into reusable skills during task execution, reducing the token consumption of subsequent repeated requests.
  • High hit rate cacheThe cache mechanism improves the hit rate of duplicate requests (official data shows 90.6%), further reducing the number of API calls and costs.

How to use OpenClacky

  • Download and installVisit the OpenClacky official website https://www.openclacky.com/ or the GitHub repository to download the installation package for your system.
  • Configuration ModelSupports BYOK (with built-in key) mode, can access mainstream models such as Claude and GPT, and can integrate with CLI model services such as Alibaba Cloud Bailian.
  • Input taskSimply enter natural language commands in the dialog box, and the Agent will automatically understand and execute them, such as generating code, scraping web page data, and creating PPT presentations.
  • Binding communication toolsIn the settings, you can connect to Lark, WeChat, DingTalk, or Discord, and the results of the task can be automatically pushed to the designated channel after completion.
  • Use SkillLoad the required skill packs from the Skill Marketplace, or let the Agent automatically accumulate frequently used operations as custom skills during tasks.
  • Scheduled hostingSet up a scheduled task (such as searching for intelligence on the X platform every morning), and the Agent will automatically wake up the browser in the background to complete the task and push the results.

OpenClacky's core advantages

  • Ultimate Savings TokenThe total cost is only 1/6 of Hermes Agent, 1/3 of OpenClaw, and lower than Claude Code, and it can be used 24/7 without any psychological burden.
  • Zero-barrier installationOne-click download, double-click to use, say goodbye to Docker, Gateway and cumbersome configuration, truly out of the box and ready to use.
  • Model switchingBYOK routing strategy allows for the use of strong models for complex tasks and cheap models for simple tasks, enabling fine-grained cost control.
  • Skill EcosystemIt supports packaging industry experience into marketable skills, enabling professionals to become micro SaaS founders with zero code.
  • Self-evolution abilityBuilt-in dual-hook mechanism automatically accumulates high-frequency operations during task execution, making it more and more convenient to use.
  • Full-modal outputA single sentence can generate various content formats, including images, videos, code, and tables, enabling you to work or create anytime, anywhere.

OpenClacky project address

  • Project official websitehttps://www.openclacky.com/
  • GitHub repositoryhttps://github.com/clacky-ai/openclacky

Comparison of OpenClacky and similar competing products

Comparison Dimensions OpenClacky Claude Code OpenClaw
Total cost of the same task $5.10(lowest) $5.49 $15.70 (3 times the price of OpenClacky)
Number of requests 51(least) 70 81
Core tool count 16(streamline) 40+ (huge) 23
Cache hit rate 90.6% 95.2%(Highest) 88.7%
Installation threshold One-click installation, zero configuration Development environment configuration required Requires Docker/gateway, high barrier to entry.
Token Strategy Streamlined tools + context compression + model routing Strong model native driver The toolset is comprehensive, but the schema is heavy.
Open source positioning First in the province: Token + Skill ecosystem Official Coding Assistant General open-source agent framework

Application scenarios of OpenClacky

  • Software developmentSimply input natural language commands to automatically build a complete front-end and back-end project, supporting code generation, debugging, and iteration, significantly reducing the development threshold.
  • Content creation and multimodal generationGenerate images, adjust styles, create posters and add text with a single sentence, and even directly output videos, meeting all your text, image, and video creation needs.
  • Web page data scraping and processingAutomatically launches the built-in browser to access the target webpage, captures key data, and organizes it into a structured table, replacing tedious manual copying and pasting.
  • Timely intelligence gatheringSet up daily scheduled tasks to automatically search for cutting-edge intelligence and industry trends on platforms such as X, continuously providing information support for decision-making.
  • Office automation and communication integrationUpon completion of the task, the results will be automatically pushed to Lark, WeChat, DingTalk, or Discord, achieving a fully unattended "execution-delivery-notification" process.