Overview
OpenClacky is the most Token-efficient open-source AI Agent. Built for developers, creators, and teams who use AI Agents heavily, it tackles a practical problem: Agents can get work done, but long and multi-turn tasks burn money. OpenClacky's core value is not "yet another AI assistant"—it uses an engineered architecture to cut Agent usage cost while keeping near Claude Code-level task execution. It ships as open-source, local-run, BYOK: bring your own API Key and connect Claude, GPT, DeepSeek, Kimi, Gemini, Minimax, OpenRouter, and other OpenAI-compatible APIs without locking to one model or platform. Use a strong model for complex work and cheaper models for ordinary sub-tasks—you decide quality vs cost. In engineering, OpenClacky specializes around Prompt Cache: freezing the system prompt, dual cache markers, Insert-then-Compress, compress-on-idle, stable tool Schema, and related mechanisms raise cache reuse in long sessions and cut repeated input Token spend.
Connected profiles
More Coding assistants products
Sources · 0Sources & provenance
English text is maintained separately from the original record. Consult the original source before citing a fact.
No source links are available for this profile.
Original source text
Names may use phonetic renderings when no English name is supplied. The original wording is retained below.
OpenClacky 是最省 Token 的开源 AI Agent。它面向高频使用 AI Agent 的开发者、创作者和团队，解决一个很现实的问题：Agent 能做事了，但长任务、多轮任务太烧钱。OpenClacky 的核心价值不是“又一个 AI 助手”，而是用工程化架构把 Agent 的使用成本打下来，同时保持接近 Claude Code 级别的任务执行能力。OpenClacky 采用开源、本地运行、BYOK 的产品形态，用户可以自带 API Key，接入 Claude、GPT、DeepSeek、Kimi、Gemini、Minimax、OpenRouter 等 OpenAI-compatible API，不被单一模型或平台绑定。复杂任务可以用强模型，普通子任务可以切换到更经济的模型，由用户自己决定效果和成本。在工程实现上，OpenClacky 围绕 Prompt Cache 做了专门优化，通过冻结系统提示词、双缓存标记、Insert-then-Compress、空闲时压缩、稳定工具 Schema 等机制，提高长会话中的缓存复用率，减少重复输入 Token 消耗。