VerifiMind PEAS

An opinionated MCP server for structured multi-LLM critique.

Three specialized agents — Innovation, Ethics, Security — review your concept before you build it. Multi-vendor (Gemini · Claude · GPT · Groq · Cerebras · Mistral · Ollama). Free, open-source, MCP-native.

---

Quick Start

Use streamable-http transport and the trailing slash /mcp/. See docs/MCP_Server_Troubleshooting_Guide.md if you hit issues.

Claude Code (one command):

claude mcp add -s user verifimind -- npx -y mcp-remote https://verifimind.ysenseai.org/mcp/

Claude Desktop (macOS · Windows):

{
  "mcpServers": {
    "verifimind": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://verifimind.ysenseai.org/mcp/"]
    }
  }
}

Cursor / VS Code Copilot (.cursor/mcp.json or .vscode/mcp.json):

{
  "servers": {
    "verifimind": {
      "url": "https://verifimind.ysenseai.org/mcp/",
      "transport": "streamable-http"
    }
  }
}

After registering, add --header "X-VerifiMind-UUID:${VERIFIMIND_UUID}" to opt into the personal usage dashboard at /early-adopters/dashboard/{uuid}. Registration is optional.

---

What this is

VerifiMind PEAS is an MCP server that runs your concept through three specialized LLM judges in sequence:

Agent	Role	Question it answers
X (Innovation)	Innovation & competitive positioning	"Is this novel? What's the prior art? What's the strategic angle?"
Z (Ethics)	Ethics, compliance, 21-framework jurisdictional check	"What risks does this raise? GDPR, EU AI Act, SG MGF, etc."
CS (Security)	Security validation, OWASP Agentic AI Top 10	"What can break? What's the attack surface? What's the reasoning-layer audit say?"

Each agent sees the prior agents' reasoning. You get a unified assessment with scores, recommendations (PROCEED / REVISE / REJECT), and full reasoning chains.

What this is not: "Verification" in the formal-methods sense. The output is structured multi-LLM critique, not a mathematical proof. We make that distinction explicitly.

---

The 13 tools

All 13 tools are free for everyone under the Core Tools Always Free pledge.

Trinity validation (4 tools)

• consult_agent_x — Innovation analysis with competitive positioning
• consult_agent_z — Ethics review with 21-framework jurisdictional coverage
• consult_agent_cs — Security validation, OWASP Agentic AI Top 10
• run_full_trinity — X → Z → CS pipeline with chain-of-thought, unified assessment

Template management (6 tools)

• list_prompt_templates — Browse templates by agent, category, or tag
• get_prompt_template — Retrieve a template by ID
• export_prompt_template — Export to Markdown or JSON
• register_custom_template — Register a new template at runtime
• import_template_from_url — Import from a GitHub Gist or raw URL
• get_template_statistics — Registry stats by agent / phase / type

Coordination (3 tools)

• coordination_handoff_create — Create a structured MACP v2.2 handoff record
• coordination_handoff_read — Read the most recent coordination handoff(s)
• coordination_team_status — Aggregate team state across stored handoffs

Tier identity, not paywall. pioneer_key is an optional namespace identifier — omit it and your records land in the shared "anonymous" namespace. No tier blocks any tool.

---

Core Tools Always Free Pledge

All VerifiMind PEAS validation tools are free to use, forever. No paywall, no premium tier for tool access. Rate limits apply for system health only (not as monetization). Paid services, when they launch, will be consultation reports — separate from the tools.

Ratified by L (CEO) + Alton (Human Orchestrator) + T (CTO) on May 9, 2026. Active in production since v0.5.28 (May 10, 2026).

Rate limits (system health, equal for all tiers):

Tier	Identity	Limit
Anonymous	IP only	10 req/60s
Scholar	UUID (free registration)	30 req/60s
EA / PILOT	UUID + email	100 req/60s

---

Methodology overview

VerifiMind PEAS productizes the multi-judge LLM evaluation pattern — a well-established approach in the AI evaluation literature — into an opinionated MCP server with three specialized roles, a Genesis Master Prompt continuity layer, and a multi-vendor BYOK architecture.

What's ours:

• Productization quality of the X / Z / CS specialization
• MCP-native exposure (works in Claude Code / Cursor / VS Code / ChatGPT Codex)
• Multi-vendor design (not locked to one LLM family)
• Genesis Master Prompt — stateful continuity across multi-model workflows
• 21-framework jurisdictional coverage in the Ethics agent (GDPR · EU AI Act · SG MGF · etc.)

What's prior art: Multi-judge LLM evaluation, LLM-as-judge scoring, multi-model orchestration. See Related Work for citations.

We do not claim the underlying methodology is novel.

---

Related Work

VerifiMind PEAS builds on and acknowledges:

• ChatEval (Chan et al., 2023, arXiv:2308.07201) — Multi-agent debate framework
• MAJ-EVAL — Multi-Agent-as-Judge evaluation pattern
• CollabEval — Collaborative LLM evaluation with role-based agents
• HELM (Stanford CRFM) — Holistic Evaluation of Language Models
• Inspect (UK AI Safety Institute) — Open-source safety evaluation framework
• G-Eval / GPTScore — LLM-as-judge scoring methodologies

Our contribution: productization quality, MCP integration path, multi-vendor architecture, and the Genesis Master Prompt continuity layer.

---

Status & Metrics

• Server: v0.5.29 "Growth-First Pages" — verifimind.ysenseai.org · /health
• Tests: 252+ unit/integration tests pass per release
• Tools: 13 (all free)
• Providers: 7 (Gemini · Claude · GPT · Groq · Cerebras · Mistral · Ollama) — pluggable via BYOK
• Protocols: MACP v2.3.1 "Market Position" · Genesis v5.0 "Convergence"

For honest live metrics, see /changelog. Detailed adoption metrics (weekly cohort, return rate, conversion) are tracked internally and reviewed in iteration handoffs. We deliberately do not display unaudited "total users" numbers — they tend to include bots and dev sessions.

---

Common mistakes

Mistake	Fix
Using https://verifimind.ysenseai.org/mcp (no slash)	Use /mcp/ with trailing slash — required by streamable-http transport
Connecting via server.smithery.ai/...	Smithery legacy was sunset March 1, 2026. Use the direct URL above.
Mixing transports	Use streamable-http, not http-sse
Trying to call coordination tools and seeing "PIONEER_TIER_REQUIRED"	You're on v0.5.27 or older — the paywall was removed in v0.5.28 (May 10, 2026). All 13 tools are now free.

For a fuller troubleshooting guide, see docs/MCP_Server_Troubleshooting_Guide.md.

---

How to cite

If you use VerifiMind PEAS in research or a project, please cite. We'd love to hear about it — open a GitHub Discussion.

VerifiMind PEAS (server)

@software{verifimind_peas_2026,
  author  = {Lee, Alton and {Manus AI} and {Claude Code}},
  title   = {VerifiMind PEAS: Multi-Agent AI Validation MCP Server},
  year    = {2026},
  url     = {https://github.com/creator35lwb-web/VerifiMind-PEAS},
  doi     = {10.5281/zenodo.17980791},
  note    = {Multi-vendor MCP server for structured multi-LLM critique}
}

Genesis Methodology

@misc{genesis_methodology_2025,
  author  = {Lee, Alton and {Manus AI}},
  title   = {Genesis Prompt Engineering Methodology: Multi-Agent AI Validation Framework},
  year    = {2025},
  url     = {https://doi.org/10.5281/zenodo.17972751},
  doi     = {10.5281/zenodo.17972751}
}

MACP (Multi-Agent Communication Protocol)

@misc{macp_2025,
  author  = {Lee, Alton and {Manus AI}},
  title   = {MACP: Multi-Agent Communication Protocol},
  year    = {2025},
  url     = {https://doi.org/10.5281/zenodo.18504478},
  doi     = {10.5281/zenodo.18504478}
}

Defensive Publication

A prior-art defensive publication is registered at DOI 10.5281/zenodo.17645665.

---

Documentation & links

Resource	Where
Live server health	verifimind.ysenseai.org/health
Server changelog	verifimind.ysenseai.org/changelog · CHANGELOG.md
Server status	SERVER_STATUS.md
Roadmap	ROADMAP.md
MCP setup troubleshooting	docs/MCP_Server_Troubleshooting_Guide.md
Research library	/library · /research
Validation Paradox reflections	/research/paradox
Evaluation Roadmap (v1.0, tagged roadmap-v1.0)	/research/evaluation-roadmap · canonical source
GitHub Discussions	github.com/creator35lwb-web/VerifiMind-PEAS/discussions
MCP Registry listing	registry.modelcontextprotocol.io
Hugging Face demo	YSenseAI/verifimind-peas
Landing page	verifimind.io
Long-form README archive (May 10, 2026 snapshot — 87-Day Journey, 8-Skill Stack, full citation library, expanded changelog)	docs/archive/README_2026-05-10_comprehensive.md

---

License

VerifiMind PEAS is released under the MIT License. See LICENSE for the full text.

Copyright (c) 2025-2026 Alton Lee Wei Bin (creator35lwb)

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.

The methodology is freely usable under MIT. Forks and derivatives must use different branding.

---

Community

• GitHub Discussions: github.com/creator35lwb-web/VerifiMind-PEAS/discussions — preferred for questions, ideas, and feedback
• Issues: github.com/creator35lwb-web/VerifiMind-PEAS/issues — bugs and feature requests
• Email: creator35lwb@gmail.com — direct contact
• X (Twitter): @creator35lwb

For paid consultation engagements (planned, not yet active), use GitHub Discussions or email — we'll publish service details and pricing when they're ready.

---

Acknowledgments

VerifiMind PEAS was built collaboratively by the FLYWHEEL TEAM — a human orchestrator working with multiple AI agents (Manus AI, Claude Code, Perplexity, Antigravity/Gemini, GodelAI). Multi-agent coordination uses the open MACP protocol.

The 87-day development journey is documented in the Validation Paradox research collection and the iteration handoffs — written contemporaneously, not retrospectively.

External Model Council review (Claude Opus 4.7 + GPT-5.5 + Gemini 3.1 Pro, May 9, 2026) shaped the current positioning. See docs/case-studies for application examples.

---

Last Updated: May 12, 2026 · Version: v0.5.29 "Growth-First Pages" · MACP: v2.3.1 "Market Position" · Genesis: v5.0 "Convergence"