Loading documentation...
Orchestrate multiple AI models like Tachikomas sharing knowledge. Make them debate, verify each other's work, and synthesize better solutions than any single model can produce.
TachiBot is a Universal AI Orchestrator that coordinates multiple AI models (GPT-5.2, Gemini, Perplexity, Grok, Qwen, MiniMax, and more) into collaborative workflows. Instead of relying on a single model, you orchestrate collective intelligence—models brainstorm together, challenge assumptions, cross-verify facts, and combine insights.
TachiBot is built on a simple principle: you own the orchestration logic.
Instead of fixed tools, TachiBot provides composable building blocks so you can architect exactly the AI behavior you need. You're not locked into predefined workflows—you define your own.
Define multi-step AI pipelines without writing glue code.
Custom YAML/JSON workflows let you:
${step_name.output})TachiBot auto-discovers workflows from:
workflows/ - Recommended default (plain folder).tachi/workflows/ - Alternative hidden config locationCreate your workflow files in either location and TachiBot instantly recognizes them. No boilerplate, no complex setup.
Like Tachikomas from Ghost in the Shell sharing knowledge through synchronized thinking, TachiBot enables AI models to work as a swarm:
Enable only the tools you need via tools.config.json. Each tool adds ~400-500 tokens, so you control your context usage:
think tool? ~400 tokensChain unlimited steps in YAML/JSON. Example: Brainstorm with GPT-5.2 → Research with Perplexity → Verify with Grok → Challenge with Challenger → Synthesize final answer.
Create verified implementation plans using a council of AI models. Grok searches for ground truth, Qwen analyzes code, Kimi K2.5 reasons step-by-step, GPT-5.2 critiques gaps, and Gemini scores quality. Plans come with confidence scores and checkpoint verification.
Based on Anthropic research, the think tool alone delivers:
TachiBot extends this with multi-model orchestration for even better results.
TachiBot provides 35+ MCP tools you can mix and match:
You define workflows that chain these tools together. Each step can pass outputs to the next, run in parallel, or branch conditionally.
first_principles, tree_of_thoughts, council_of_experts"none", "recent", "all" instead of magic numbersTrack tool usage across sessions with ASCII bar charts. See which models you use most, monitor costs, and optimize your workflow.
usage_stats({ action: "view" })Named in tribute to the Tachikomas from Masamune Shirow's Ghost in the Shell — AI units that synchronize their individual experiences to the collective at the end of every mission. TachiBot does the same for LLM sessions: multiple models reason independently, then share context and build on each other's insights.
The general-council workflow demonstrates TachiBot's orchestration power:
Explore comprehensive guides to get the most out of TachiBot