DOCS BLOG ABOUT TRY THE GATEWAY

// MCP TOOL REFERENCE

TACHIBOT TOOLS

104 tools from the Tachibot MCP Server, categorised by risk level.

View the Tachibot policy →

// ALL 104 TACHIBOT TOOLS

READ 94 tools

Read ai Query any AI model Read ai_analyze Analyze code or text Read ai_code Generate code Read architect Full codebase analysis. Put your QUERY in the Read auditor Evidence-based audit. Put the CONTEXT in the Read basic-response Return basic error information Read brainstorm Multi-model brainstorming session Read cached-results Return previously cached successful results Read cheaper-model Fallback to GPT-5 Nano for cost efficiency Read code_reviewer Code review. Put the CODE in the Read code-review Comprehensive code review with multiple perspectives Read continue_focus Continue a focus session Read continue_workflow Continue workflow Read cyberpunk High contrast neon on black - Blade Runner vibes Read debug-assistant Multi-stage debugging helper Read devlog Session logging with devlog-mcp (workspace tracking, daily logs) Read diagnostic_test Test diagnostic tool Read documentation_writer Documentation generation. Put the CODE in the Read focus Multi-model reasoning Read focus_analyze_sessions Analyze multiple sessions for patterns, model performance, and insights Read focus_find_similar Find sessions similar to a given query for learning from past interactions Read focus_list_sessions List all saved Focus MCP brainstorming sessions with metadata Read focus_recommend Get mode and model recommendations based on session history Read focus_replay_session Replay a saved session to see all model responses and synthesis Read focus_session_status Get the status of the current active session Read gemini_analyze_code Analyze code for bugs, quality, security, or performance issues. Put the CODE in the Read gemini_analyze_text Rhetorical analysis: dissect arguments for bias, logical fallacies, and persuasion tactics. Use for evaluat... Read gemini_brainstorm Convergent synthesis: cluster, refine, and prioritize raw ideas into structured hierarchies. Use AFTER dive... Read gemini_image_prompt Image prompt generation. Put the DESCRIPTION in the Read gemini_judge Evaluate and synthesize multiple AI perspectives into a unified verdict. Put CONTENT in the Read gemini_query Query Gemini. Put your PROMPT in the Read gemini_search Web search via Gemini with Google Search grounding Read gemini_summarize Summarization. Put the CONTENT in the Read gpt5_mini_reason Cost-efficient reasoning using GPT-5-mini. Put your QUERY in the Read gpt5_reason Advanced reasoning using GPT-5. Put your QUERY in the Read grok_architect Architecture design. Put your REQUIREMENTS in the Read grok_brainstorm Contrarian first-principles brainstorming: deconstruct a topic to atomic truths, challenge every assumption... Read grok_code Code analysis. Put the CODE in the Read grok_debug Debug assistance. Describe the ISSUE in the Read grok_function Use Grok-4 Read grok_reason Deep reasoning. Put your PROBLEM or QUESTION in the Read grok_reason_v4 Deep reasoning Read grok_scout Quick research Read grok_search Web search Read jury Multi-model jury: runs question through configurable panel of AI jurors in parallel, then Gemini synthesize... Read kimi_code SWE-focused code generation/fixing with Kimi K2.5 (SWE-Bench 76.8%). Put your REQUEST in the Read kimi_decompose Structured task decomposition with Kimi K2.5 Agent Swarm reasoning. Breaks tasks into subtasks with IDs, de... Read kimi_long_context Long-context analysis with Kimi K2.5 (256K context window). Put CONTENT in the Read kimi_thinking Kimi K2.5 multimodal reasoning with Agent Swarm. Put your PROBLEM in the Read legacy Legacy flat configuration Read list_plans List recently created plans from planner_maker. Shows plans from the last N days (default 7) with filename,... Read list_prompt_techniques Discover available prompt engineering techniques. Shows all 31 techniques organized by category. Read list_workflows List workflows Read lmstudio_chat Have a conversation with local models. Put your MESSAGE in the Read lmstudio_code Generate or analyze code. Put REQUIREMENTS in Read lmstudio_creative Creative writing and ideation. Put your PROMPT in the Read lmstudio_query Query local open-weight models via LM Studio Read lmstudio_reason Perform step-by-step reasoning using local models. Put your PROBLEM in the Read mem0 AI memory with semantic search via mem0 Read minimal Clean Swiss typography, minimal color usage Read minimax_agent Multi-step task decomposition and execution with MiniMax M2.7: plan, analyze, research, decide. Use when a ... Read minimax_code Single-pass code operations with MiniMax M2.7 (SWE-Pro 56.22%, #1 AI Intelligence Index). Put your REQUEST ... Read nebula Modern SaaS look with soft pastels (default) Read nextThought Sequential thinking with optional multi-model execution. Auto-creates session if needed. **Basic** (though... Read ocean Cool blues and teals - calm and professional Read openai_brainstorm Find alternative approaches: when stuck on a programming problem, reveals 3rd/4th/5th options you haven Read openai_code_review Code review. Put the CODE in the Read openai_explain Explain concepts. Put the TOPIC in the Read openai_reason Mathematical reasoning using GPT-5.2-thinking. Put your QUERY in the Read openai_search Web search using GPT-5.2 with real-time web access. Put your QUERY in the Read openrouter_multi Access Qwen and Kimi models. Put your QUERY in the Read perplexity_ask Web search. Put your QUERY in the Read perplexity_code_search Code search. Put your QUERY in the Read perplexity_fact_check Fact-check claims. Put the CLAIM in the Read perplexity_reason Reasoning with search. Put your PROBLEM in the Read perplexity_research Deep research using sonar-deep-research. Synthesizes hundreds of sources into a comprehensive report. Put y... Read preview_prompt_technique Preview how a technique enhances your prompt WITHOUT executing. Returns an execution_token for later use. Read qwen Access Qwen3-Coder, QwQ reasoning, and other OpenRouter models. Select Read qwen_algo Expert algorithm analysis: complexity profiling, optimization tiers, constraint-driven recommendations, com... Read qwen_coder Code generation and analysis with Qwen3-Coder-Next. Put your REQUEST in the Read qwen_competitive Competitive programming. Put the PROBLEM in the Read qwen_general General-purpose assistance with Qwen3. Put your QUERY in the Read qwen_reason Heavy mathematical reasoning with Qwen3-Max-Thinking (>1T params, 98% HMMT). Put your PROBLEM in the Read qwq_reason Multi-perspective deliberation: simulate 4 opposing viewpoints (optimist/pessimist/domain-expert/contrarian... Read reduced-scope Retry with reduced complexity or scope Read simpler-tool Use a simpler, more reliable tool Read tachi Smart AI assistant - just describe what you need. Put your QUERY in the Read test Test tool Read test_architect Test suite design. Put the CODE in the Read think Log a thought Read validate_workflow Validates workflow YAML/JSON files for correctness. Checks: - Valid YAML/JSON syntax - Interpolation refer... Read validate_workflow_file Validates a workflow file from the filesystem. Checks: - Valid YAML/JSON syntax - Interpolation references... Read visualize_workflow Visualize workflow Read workflow_status Workflow status

WRITE 5 tools

Write commit_guardian Pre-commit validation. Put the CONTEXT in the Write create_workflow Create a new workflow. Put the WORKFLOW NAME in the Write focus_export_session Export a session in different formats for sharing or archiving Write planner_maker Multi-model council for creating implementation plans. COORDINATOR PATTERN - Returns ONE tool to execute a... Write planner_runner Execute implementation plans step-by-step with goal-oriented verification. COORDINATOR PATTERN - tracks ac...

DESTRUCTIVE 2 tools

Destructive focus_clear_old_sessions Clean up old session files to save disk space Destructive usage_stats View or reset tool usage statistics. Put your REQUEST in the

EXECUTE 3 tools

Execute execute_prompt_technique Execute a prompt technique. Use execution_token from preview, OR provide full params (technique, tool, quer... Execute workflow Execute workflows with Ink rendering, comparison table, and optional AI judge Execute workflow_start Start workflow session

// RUNNING THIS SERVER

The managed route: connect Tachibot through the PolicyLayer gateway — every tool call above is checked against your policy before it runs, with a full audit log.

DIRECT INSTALL (UNMANAGED) npx -y tachibot-mcp

// FAQ

How many tools does the Tachibot MCP server have? +

The Tachibot MCP server exposes 104 tools across 4 categories: Read, Write, Destructive, Execute.

How do I enforce policies on Tachibot tools? +

Route the Tachibot server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do Tachibot tools fall into? +

Tachibot tools are categorised as Read (94), Write (5), Destructive (2), Execute (3). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.