TachiBot/Docs
HomeGitHubnpm
IntroductionInstallationQuick StartExamplesTools OverviewPlannerPrompt TechniquesWorkflowsPingPongTool ProfilesTools ConfigThemesUsage StatsAPI KeysWhat's NewTerms & ConditionsCookie Policy
Products
TachiBot MCPdokorotachi-agent
Getting started
  • Introduction
  • Installation
  • Quick Start
  • Examples
Core features
  • Tools Overview
  • Planner
  • Prompt Techniques
  • Workflows
  • PingPong
Configuration
  • Tool Profiles
  • Tools Config
  • Themes
  • Usage Stats
  • API Keys
Resources
  • What's New
Legal
  • Terms & Conditions
  • Cookie Policy

Loading documentation...

← Previous
Installation
Next →
Examples
Edit this page on GitHub
Docs/Getting started

Quick Start

Get running in 5 minutes. Learn the basics and run your first multi-model workflow.

Your first workflow

Run a workflow that uses multiple models to verify an answer.

terminalshell
1tachibot workflow run general-council "What are the best practices for REST API design?"
  1. 1
    Brainstorm
    Initial ideas with GPT-5.6.
  2. 2
    Research
    Best practices with Perplexity.
  3. 3
    Verify
    Cross-check with Grok.
  4. 4
    Synthesize
    The final answer.

PingPong debates

Make models debate and build on each other’s ideas.

terminalshell
1tachibot pingpong \
2 --rounds 5 \
3 --style collaborative \
4 "Design a scalable microservices architecture"

Configure your tools

Control which tools load to manage token usage.

tools.config.jsonjson
1{
2 "tools": {
3 "think": true,
4 "focus": true,
5 "perplexity_ask": true,
6 "grok_reason": false,
7 "gemini_brainstorm": true
8 }
9}
✓
Pro tip
Start with fewer tools enabled, then add more as needed. Each tool adds about 250 tokens at the median.

Build a custom workflow

my-workflow.yamlyaml
1name: api-verification
2description: Verify API design with multiple models
3
4steps:
5 - tool: openai_brainstorm
6 params:
7 problem: "Design REST API for user authentication"
8 quantity: 3
9 output: ideas
10
11 - tool: perplexity_reason
12 params:
13 query: "REST API authentication best practices"
14 output: research
15
16 - tool: plan_critique
17 params:
18 plan: "${ideas} based on ${research}"
terminalshell
1tachibot workflow run my-workflow

Common use cases

Hallucination prevention

terminalshell
1tachibot jury \
2 --jurors "perplexity,grok,openai" \
3 "Is this API endpoint design correct?"

Code review

terminalshell
1tachibot workflow run code-review \
2 --file ./src/auth.ts

Research & synthesis

terminalshell
1tachibot perplexity_reason \
2 "Latest trends in AI model orchestration"