Problem
Developers running multiple AI coding agents (Claude Code, Codex, Aider) simultaneously have no unified management interface:
- Manual context-switching between multiple terminal windows
- No unified view of agent status, current tasks, or outputs
- MCP tool provisioning (search, file access, DB) configured per-agent individually
- Agent output review and merge conflict resolution is fragmented
Pain Intensity: 8/10 - 80% of enterprises plan multi-agent but less than 10% successfully orchestrate
Market
- Primary Market: Power developers and small teams heavily using AI coding agents
- Segment: Claude Code, Codex, Aider, Cursor users
- TAM: AI Code Tools market $7-13B (2025), 46% CAGR
- SAM: Multi-agent orchestration tooling $150-370M
Solution
Agent Control Plane - Docker-deployable multi-agent orchestration web dashboard with integrated MCP tool provisioning
Core Features
- Agent Session Dashboard: Real-time monitoring of all AI agent status, current tasks, and outputs
- Task Assignment: Create tasks in web UI and route to specific agent sessions
- MCP Tool Provisioning: Centrally manage and auto-connect MCP tools (search, file access, DB) to agents
- Output Review UI: Review agent code changes as diffs, accept or reject
- Session State Persistence: Save and restore agent session contexts
Usage Scenario
# One-line Docker deployment
$ docker-compose up -d agent-control-plane
# Access dashboard in browser
# → Add agent sessions (Claude Code, Codex, Aider)
# → Assign MCP tools to each agent (search, GitHub, DB)
# → Create tasks → Route to agents
# → Review outputs → Accept/merge
Competition
| Competitor | Price | Weakness |
|---|---|---|
| Claude Squad | Free (OSS) | Terminal-only, no web UI, no MCP support |
| CLI Manager | TBD | Minimal features, no team support |
| VS Code Agent Sessions | Free | IDE-locked, no Docker deploy, no MCP provisioning |
| Claude-Flow | Free (OSS) | SDK/framework, no management dashboard |
Competition Intensity: Medium - No polished web dashboard product exists Differentiation: Agent-agnostic + MCP tool provisioning + team support + self-hosted
MVP Development
- MVP Timeline: 8 weeks
- Full Version: 6 months
- Tech Complexity: Medium
- Stack: React (dashboard), Node.js (backend API), Docker, WebSocket/SSE, MCP protocol
MVP Scope
- React web dashboard (agent session list, status monitoring)
- Agent session management API (create, stop, status)
- Basic task assignment workflow
- Docker Compose one-click deployment
Revenue Model
- Model: Freemium + Subscription
- Pricing:
- Free: 1 user, 2 agent sessions
- Pro: $29/mo (1 user, unlimited agents, MCP tool management)
- Team: $79/mo (5 users, team dashboard, webhook alerts)
- Self-hosted License: $299 one-time or $49/mo
- Expected MRR (6 months): $3,000-8,000
- Expected MRR (12 months): $15,000-35,000
Risk
| Type | Level | Mitigation |
|---|---|---|
| Technical | Low | React + Node.js + Docker = core skill alignment |
| Market | Medium | VS Code/Cursor may absorb natively → differentiate with self-host + agent-agnosticism |
| Execution | Medium | MCP protocol ecosystem still early → fast entry creates first-mover advantage |
Recommendation
Score: 93.5/100 ⭐⭐⭐⭐⭐
Why Recommended
- Explosive 46% CAGR growth in AI agent market
- No polished web dashboard product exists — true blue ocean
- React + Node.js + Docker = perfect skill set alignment
- Self-hosted model counters 2025 developer frustration with pricing instability
- MCP protocol native first-mover opportunity
Risk Factors
- VS Code (v1.109) already shipped Agent Sessions view → 6-12 month competitive window
- MCP ecosystem standardization uncertainty
First Actions
- Build basic agent session management PoC with Docker Compose
- Prototype Claude Code + Aider session connection via WebSocket
- Dogfood in own development workflow first
This idea combines AgentWire (tmux-based voice-control AI agents) and Google Search MCP (no-API-key search) into a unified platform for agent management + tool provisioning.