TachiBot
← Back to Home

Getting Started

  • Introduction
  • Installation
  • Quick Start
  • Examples

Core Features

  • Tools Overview
  • Planner
  • Prompt Techniques
  • Workflows

Configuration

  • Tool Profiles
  • Tools Config
  • Themes
  • Usage Stats
  • API Keys

Resources

  • What's New

Legal

  • Terms & Conditions
  • Cookie Policy

Loading documentation...

Tool Profiles

Pre-configured tool profiles for different use cases. Choose the right balance of capability and token usage for your workflow.

Quick Overview

Minimal
13 tools
~4-5k tokens
Research Power
35 tools
~9-10k tokens
Code Focus
42 tools
~9-10k tokens
Balanced
53 tools
~11-12k tokens
Heavy Coding
57 tools
~15k tokens
Full
DEFAULT
64 tools
~19-20k tokens

How to Use Profiles

Set the TACHIBOT_PROFILE environment variable or add it to your .env file:

.env

Profile Details

Minimal

Essential tools for basic tasks

13 tools
~4-5k tokens
Enabled Tools:
thinkfocustachidoctornextThoughtusage_statsperplexity_askgrok_reasongemini_brainstormqwen_coderminimax_codeworkflowlist_prompt_techniques
Best for: Budget-conscious users, simple queries, basic research and brainstorming

Research Power

Research-focused with all Perplexity + Grok search + brainstorming

35 tools
~9-10k tokens
Enabled Tools:
thinkfocustachidoctornextThoughtusage_statsperplexity_askperplexity_reasongrok_reasongrok_searchopenai_brainstormopenai_searchgemini_brainstormgemini_searchqwen_coderqwen_reasonqwq_reasonkimi_thinkingkimi_codekimi_decomposekimi_long_contextminimax_agentdeepseek_reasonglm_reasonstepfun_reasonernie_reasongemini_judgejuryworkflowlist_prompt_techniquespreview_prompt_techniqueexecute_prompt_techniqueplanner_makerplanner_runnerlist_plans
Best for: Deep research, fact-checking, academic work, complex investigations

Code Focus

Code-heavy work with debugging and analysis

42 tools
~9-10k tokens
Enabled Tools:
thinkfocustachidoctornextThoughtusage_statsperplexity_askgrok_reasongrok_codegrok_debugdebug_triageopenai_code_reviewspec_writerrefine_promptgemini_brainstormgemini_analyze_codeqwen_coderqwen_algoqwq_reasonkimi_thinkingkimi_codekimi_decomposekimi_long_contextminimax_codedeepseek_reasondeepseek_algoglm_reasonstepfun_reasonlocal_queryworkflowlist_workflowsvalidate_workflowlist_prompt_techniquespreview_prompt_techniqueexecute_prompt_techniqueplanner_makerplanner_runnerlist_planstestgensecurity_reviewdiff_reviewplan_critique
Best for: Software development, code review, debugging, refactoring

Balanced

Balanced set for general use

53 tools
~11-12k tokens
Enabled Tools:
thinkfocustachidoctornextThoughtusage_statsperplexity_askperplexity_reasongrok_reasongrok_codedebug_triagegrok_searchopenai_reasonopenai_brainstormopenai_searchspec_writerrefine_promptgemini_brainstormgemini_analyze_codegemini_analyze_textgemini_searchqwen_coderqwen_algoqwen_reasonqwq_reasonkimi_thinkingkimi_codekimi_decomposekimi_long_contextminimax_codeminimax_agentdeepseek_reasondeepseek_algoglm_reasonstepfun_reasonernie_reasonlocal_querygemini_judgejuryworkflowlist_workflowscreate_workflowvisualize_workflowlist_prompt_techniquespreview_prompt_techniqueexecute_prompt_techniqueplanner_makerplanner_runnerlist_planstestgensecurity_reviewdiff_reviewplan_critique
Best for: General purpose work, mixed code and research tasks, most users

Heavy Coding

All reasoning & code tools for intensive development

57 tools
~15k tokens
Enabled Tools:
thinkfocustachidoctornextThoughtusage_statsperplexity_askperplexity_reasongrok_reasongrok_codegrok_debugdebug_triagegrok_architectgrok_brainstormgrok_searchopenai_reasonopenai_brainstormopenai_searchopenai_code_reviewopenai_explainspec_writerrefine_promptgemini_brainstormgemini_analyze_codegemini_analyze_textgemini_searchgemini_judgejuryqwen_coderqwen_algoqwen_reasonqwq_reasonkimi_thinkingkimi_codekimi_decomposekimi_long_contextminimax_codeminimax_agentdeepseek_reasondeepseek_algoglm_reasonstepfun_reasonlocal_queryworkflowlist_workflowsworkflow_startcontinue_workflowlist_prompt_techniquespreview_prompt_techniqueexecute_prompt_techniqueplanner_makerplanner_runnerlist_planstestgensecurity_reviewdiff_reviewplan_critique
Best for: Intensive coding sessions, architecture work, full-stack development with research support

Full

DEFAULT

All tools enabled for maximum capability

64 tools
~19-20k tokens
All 64 MCP tools enabled, including:
Core (6):
think, focus, tachi, doctor, nextThought, usage_stats
Perplexity (2):
ask, reason
Grok (7):
reason, code, debug, debug_triage, architect, brainstorm, search
OpenAI (6):
reason, brainstorm, code_review, explain, search, spec_writer
Gemini (5):
brainstorm, analyze_code, analyze_text, search, judge
OpenRouter (18):
qwen_coder, qwen_algo, qwen_reason, qwen_competitive, qwq_reason, kimi_thinking, kimi_code, kimi_decompose, kimi_long_context, minimax_code, minimax_agent, deepseek_reason, deepseek_algo, glm_reason, stepfun_reason, ernie_reason, testgen, security_review
Jury, Local & Review (4):
jury, local_query (Ollama / LM Studio / llama.cpp / vLLM), diff_review, plan_critique
Workflow (9):
workflow, list, create, visualize, start, continue, status, validate, validate_file
Planner & Prompts (7):
planner_maker, planner_runner, list_plans, list/preview/execute_prompt_techniques, refine_prompt
Best for: Power users, complex workflows, maximum capability, unlimited context budgets

Create Custom Profiles

You can create your own custom profiles by creating a tools.config.json file. See the Tools Configuration page for details.

Track Your Usage

Use usage_stats to see which tools you use most, then create a custom profile with just those tools. Less tokens = faster responses and lower costs.

Learn about Usage Stats →

Related Resources

All Tools ReferenceComplete documentation for all 64 toolsTools ConfigurationCreate custom tool configurations