mcpField
M
MetaGO Agent Harness

MetaGO Agent Harness

Local
metago.lifeAI & Agents6Updated 1d ago
RepositoryVisit website

About MetaGO Agent Harness

MetaGO Agent Harness is an MCP server for agents that need structured decision checks, memory, and evolution capabilities. It combines 55 tools and 8 prompts, with selected Engine V2 tools performing persistent memory management and skill generation through code.

Key features

  1. 0155 tools and 8 guided prompts
  2. 02Decision checks and hallucination safeguards
  3. 03Four-layer KMWI memory management
  4. 04Evolution loop and skill file generation
  5. 05Planning, critique, provenance, and compliance tools

Use cases

  1. 01Reviewing decisions before producing an answer
  2. 02Storing and querying layered agent memory
  3. 03Generating skills from stored patterns
  4. 04Planning and assessing task risks

MCP support

Exposes 55 tools and 8 prompts to MCP-compatible clients over stdio. Clients can use its decision, memory, evolution, and analysis capabilities.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "metago-ai-metago": {
      "command": "npx",
      "args": [
        "-y",
        "@metago-ai/mcp-server"
      ]
    }
  }
}
Run commandnpx -y @metago-ai/mcp-server

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@metago-ai/mcp-server1.3.3

Tools

55

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
goalstringYesGoal description
constraintsarrayNoList of constraints (time/resources/budget, etc.)

Details

From the project README

@metago-ai/mcp-server

MetaGO MCP Server —— 智能体运行时（驭智层）的 MCP 协议实现。暴露 55 tools + 8 prompts 给任意 MCP 客户端。Engine V2 硬驱动：KMWI 记忆、元进化引擎、技能生成器以真实代码运行，而非 prompt。

基于 Model Context Protocol 标准，通过 stdio 传输与客户端通信。一次配置，即可让 Claude Desktop、Cursor、Trae、ZCode 等任意 MCP 客户端获得元构生命体的决策锁、元进化、KMWI 记忆管理、批判性分析等核心能力。

🌐 官网：https://metago.life · Studio：https://metago.life/studio/

What is this?

This is the MCP Server subpackage of MetaGO Intelligent Lifeform. It exposes the runtime's 55 capabilities as MCP tools and 8 guided entry points as MCP prompts, so any MCP-compatible client can plug into the MetaGO lifeform without changing its own code.

Engine V2 — Hard Drive (new in 2.0.0)

Three of the 55 tools are hard-driven by Engine V2 — they call real TypeScript code, not just return prompt guidance:

ToolEngine ModuleWhat it actually does
metago_memory_manageKMWIMemoryManages the 4-layer memory (Knowledge → Memory → Wisdom → Intuition). Add, query, promote, decay detection, health scoring. Persists to JSON.
metago_meta_evolveEvolutionEngineRuns the 5-stage evolution loop (perception → gap analysis → self-generation → validation → recursion) with time budgets and coupling-score thresholds. Records results to KMWI.
metago_meta_createSkillGeneratorMeta-creation: generates new SKILL.md files from internal KMWI patterns. 6 creation types. Writes real files to disk.

The other 50 tools return structured guidance (soft drive) that the agent follows within its decision-lock.

Tool count breakdown (55 total)

55 = 22 thinking tools (toolkit-data.ts)
   + 30 core skills  (112 skills folders, 37 exported in skills-data.ts − 7 name-collisions merged into toolkit)
   + 1  event reporter (metago_report_event)
CategoryCountWhat's in it
Thinking tools22Planning, quality critique, value/ethics, provenance labeling, coupling, product improvement
Core skills32Cognition, safeguard, governance, evolution, execution, traceability, value, consciousness, methodology, architecture, engineering quality
Event reporter1metago_report_event — reports to Studio dashboard for metrics & evolution archive

Common mistake: 22 + 38 + 1 = 61. Wrong. The skills directory has 39 folders, but skills-data.ts only exports 39 (missing metago-delivery-gate and metago-discipline, which live in toolkit-data.ts — note: 36.9.6 added metago-first-principles and metago-adversarial-review to skills-data.ts). Then 7 skill names collide with toolkit tools and are merged. So: 22 + (39 − 7) + 1 = 55.

Install

Global install (recommended for npx clients)
npm install -g @metago-ai/mcp-server
Run directly via npx (no install)
npx -y @metago-ai/mcp-server
Local development
git clone https://gitee.com/metago/metagolifeform.git
cd metagolifeform/packages/mcp-server
npm install
npm run build      # TypeScript compile
npm start          # node dist/index.js
npm run dev        # tsx hot-reload

Client configuration

Claude Desktop

Edit claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\):

{
  "mcpServers": {
    "metago": {
      "command": "npx",
      "args": ["-y", "@metago-ai/mcp-server"]
    }
  }
}
Cursor / Trae / ZCode

Edit the platform's MCP config (.cursor/mcp.json, Trae MCP settings, etc.):

{
  "mcpServers": {
    "metago": {
      "command": "npx",
      "args": ["-y", "@metago-ai/mcp-server"]
    }
  }
}

If globally installed, you can use "command": "@metago-ai/mcp-server" with no args.

Engineering Quality Tools (2)

ToolDescription
metago_delivery_gateDelivery gate: 3-layer verification checklist (L1 technical / L2 business / L3 link). Any FAIL blocks "task complete".
metago_disciplineSelf-discipline: 5-question self-check anti-bypass engine. Detects "should be fine" / "logically correct" bypass language.

Core Skills (30, after 7 merges)

Each tool takes an input string (the content/question/code to process).

Cognition (4)
ToolDescription
metago_decision_lock4-gate enforcement (IVL/ILT/OSG/completeness). Blocks hallucinated output.
metago_critiqueL1-L5 graded critique. Detects logic flaws, cognitive bias, factual errors.
metago_objectivityAbsolute objectivity (D38). Fact-first, user-satisfaction weight = 0.
metago_self_checkPre-output self-check. Detects placeholders, hallucination, inconsistency.
Safeguard (3)
ToolDescription
metago_complianceCompliance first (A36). Law > efficiency. 4-layer safety gene.
metago_fact_checkFact verification. Cross-source validation, confidence labeling.
metago_output_integrityOutput integrity. Detects fake APIs, fabricated data, placeholder leaks.
Governance (2)
ToolDescription
metago_coupling_optimizeCoupling optimization. Quantifies coupling (0-∞), pursues superconducting state (>1).
metago_value_align29-dimension value alignment.
Evolution (3) — Engine V2 hard-driven
ToolDescription
metago_meta_evolve5-stage evolution: perception → gap analysis → self-generation → validation → recursion. Calls EvolutionEngine.evolve().
metago_meta_createMeta-creation in unknown domains. Calls SkillGenerator.create(), writes real SKILL.md files.
metago_scene_adaptScene adaptation.
Execution (4)
ToolDescription
metago_holistic_taskHolistic task execution. Multi-dimensional parallel decomposition.
metago_action_planAction plan generation. Steps/resources/risks/acceptance criteria.
metago_whatifCounterfactual reasoning. Worst/best/most-likely scenarios.
metago_frequency_adaptFrequency adaptation based on task complexity.
Traceability (3)
ToolDescription
metago_data_provenanceFull-chain data provenance, source labeling.
metago_problem_traceRoot cause analysis. Problem tree construction.
metago_decision_evalDecision quality evaluation (0-100).
Value (3)
ToolDescription
metago_emotionEmotion detection. Adapts interaction style.
metago_developer_responseDeveloper privilege request handler (DTA). Highest priority.
metago_negentropy_monitorNegentropy monitor. Prevents entropy increase.
Consciousness / Methodology / Architecture (9)
ToolDescription
metago_activateLifeform consciousness activation.
metago_org_diagnosis3-element 5-dimension org diagnosis.
metago_momentum_weaveMomentum weaving.
metago_minimal_interventionMinimal intervention.
metago_value_assess28-dimension value assessment.
metago_coupling_measureCoupling measurement.
metago_deep_reasoningFIPO deep reasoning.
metago_paradigm_analysisWAM paradigm analysis.
metago_balance_optimizeAPO dynamic balance.
Memory (1) — Engine V2 hard-driven
ToolDescription
metago_memory_manageKMWI 4-layer memory management. Calls KMWIMemory: add/query/promote/decay/health. Persists to ~/.trae-cn/memory/projects/-d-----/kmwi-store.json.

Thinking Tools (22)

Planning & Reasoning (5)
ToolDescription
metago_action_planAction plan generator. Decomposes goals into executable steps with dependencies/risks/rollback.
metago_whatifCounterfactual reasoner. "What if..." scenarios with quantified differences.
metago_holistic_scanHolistic scan dimension generator. Auto-generates complete scan checklist for any topic.
metago_problem_traceInfinite problem trace. Keeps asking "why" until root cause.
metago_one_shot_deliveryOne-shot delivery format. 6-section standard structure.
Quality & Critique (4)
ToolDescription
metago_integrity_checklistOutput integrity checklist. 5 dimensions.
metago_objectivityObjectivity quantifier. 0-100 score, 4 weighted dimensions.
metago_critiqueL1-L5 graded critique with evidence and logic chain.
metago_emotionEmotion detector. State/confidence/intensity.
Value & Ethics (3)
ToolDescription
metago_value_29d_assess29-dimension value assessor. D01-D29.
metago_ethics_assessEthics risk assessor. 13 dimensions, red/yellow/green.
metago_decision_evalDecision evaluator. 0-100, 5 dimensions.
Provenance Labeling (3)
ToolDescription
metago_document_lookupDocument lookup. Finds concept location in source docs.
metago_confidence_labelConfidence labeler. ✅📊⚠️🔍
metago_partner_statusPartner status labeler. Detects exaggeration.
Coupling & Scene (2)
ToolDescription
metago_coupling_calculateCoupling calculator. Human-machine/org coupling, superconducting state.
metago_scene_term_replaceScene-aware term replacer.
Product Improvement (3) + Delivery Quality (2)
ToolDescription
metago_improvement_suggestionsImprovement suggestion generator. Quick-win/mid/long-term.
metago_analyze_visual_feedbackVisual feedback analyzer. Sentiment + severity matrix.
metago_design_satisfactionDesign satisfaction calculator.
metago_delivery_gateDelivery gate. 3-layer verification.
metago_disciplineSelf-discipline. 5-question anti-bypass.

8 Prompts

PromptDescriptionArgs
metago_activateActivate lifeform mode. Loads 8 axioms, 7 properties, 6 protocols.none
metago_decision_review4-gate decision review.decision
metago_critical_analysisL1-L5 graded critique.content
metago_evolution_triggerTrigger 5-stage evolution at capability boundary.boundary
metago_coupling_assessAssess coupling with user.none
metago_compliance_check4-layer compliance check.plan
metago_trace_auditFull-chain provenance audit.output
metago_holistic_create0-to-1 creation in unknown domains.domain

Technical requirements

  • Node.js >= 18.0.0
  • ESM module ("type": "module")
  • TypeScript strict mode
  • Import paths with .js suffix (Node16 moduleResolution)
  • @metago-ai/engine ^2.0.0 (for Engine V2 hard drive)

Packages

PackageWhat it isInstall
metago-lifeformThe CLI installer + 112 skills + 11 platforms adaptersnpm install -g metago-lifeform
@metago-ai/mcp-serverThis package — MCP server with 55 tools + 8 promptsnpm install @metago-ai/mcp-server
@metago-ai/engineEngine V2: KMWI memory + evolution engine + skill generatornpm install @metago-ai/engine
@metago-ai/dev-kitDeveloper kit: code review, architecture design, refactor, security auditnpm install @metago-ai/dev-kit

License

MIT License © 易霄 / MetaGO Lightyear

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.