Tier-1380 MCP Integration Skill

Version: 1.0.0
Tier: 1380%
Domain: factory-wager.com
Protocol: Model Context Protocol (MCP) 2024-11-05

Overview

This skill provides comprehensive MCP (Model Context Protocol) integration for the Tier-1380 ecosystem:

• MCP Server Management - Configure, start, and monitor MCP servers
• Tool Orchestration - Unified interface for OpenClaw, Matrix Agent, and infrastructure
• AI Assistant Integration - Claude, Kimi, and other MCP-compatible assistants
• Shell Bridge - Connect shell commands with AI context
• Flow Automation - Automated workflows combining multiple MCP tools

Quick Commands

MCP Server Control

# List MCP servers
bun run mcp:list                      # List configured servers
bun run mcp:status                    # Check all server status
bun run mcp:start <name>              # Start specific server
bun run mcp:stop <name>               # Stop specific server
bun run mcp:restart <name>            # Restart server

# Inspect MCP
bun run mcp:inspector                 # MCP inspector UI

Unified Shell Bridge

# Via unified-shell-bridge.ts
bun run mcp:shell                     # Start shell bridge MCP

# Direct usage
bun ~/.kimi/tools/unified-shell-bridge.ts

Bun Docs MCP

# Run Bun docs MCP server
bun run mcp:bun-docs                  # Start Bun docs server
bun run mcp:bun-test                  # Test Bun MCP integration

# Search Bun docs
bun run docs:search "URLPattern"      # Search documentation
bun run docs:entry "Bun.file"         # Get curated entry

MCP Server Configuration

Server Registry (.mcp.json)

Unified Shell Bridge Tools

The unified-shell-bridge.ts provides these MCP tools:

MCP Flows

Flow 1: OpenClaw Health Check

name: openclaw-health
steps:
  1. openclaw_status:
     - Check gateway status
     - Get version and latency
  2. matrix_agent_status:
     - Verify agent is running
     - Check configuration
  3. shell_execute:
     - Run: bun run matrix:openclaw:health
     - Parse health report
  4. decision:
     - if all healthy: report success
     - if degraded: attempt restart
     - if down: alert + escalate

Execute:

bun ~/.kimi/skills/tier1380-mcp/scripts/mcp-flow.ts openclaw-health

Flow 2: Profile Deployment

name: profile-deploy
steps:
  1. profile_list:
     - Get available profiles
  2. profile_switch:
     - Activate target profile
  3. shell_execute:
     - Validate environment
     - Run: bun run matrix:profile:use <name> --dry-run
  4. openclaw_status:
     - Verify gateway accepts new profile
  5. shell_execute:
     - Apply profile: eval $(bun run matrix:profile:use <name>)
  6. profile_status:
     - Confirm terminal binding updated

Execute:

bun ~/.kimi/skills/tier1380-mcp/scripts/mcp-flow.ts profile-deploy <name>

Flow 3: Full System Diagnostic

name: system-diagnostic
steps:
  1. parallel:
     - openclaw_status
     - matrix_agent_status
     - shell_execute: infra status
     - shell_execute: omega registry check
  2. sequential-thinking:
     - Analyze all component statuses
     - Identify issues and root causes
  3. shell_execute:
     - Run: bun run matrix:openclaw:status --json
  4. github:
     - Create issue if critical issues found
  5. report:
     - Generate diagnostic report
     - Send notification

Execute:

bun ~/.kimi/skills/tier1380-mcp/scripts/mcp-flow.ts system-diagnostic

Flow 4: Legacy Migration

name: legacy-migration
steps:
  1. clawbot_legacy_config:
     - Read legacy clawdbot.json
  2. shell_execute:
     - Check if already migrated: ls ~/.matrix/.migrated-from-clawdbot
  3. clawbot_migrate:
     - Run migration process
  4. matrix_agent_status:
     - Verify new agent works
  5. openclaw_status:
     - Verify gateway connection
  6. shell_execute:
     - Test: bun run matrix:openclaw:health

Execute:

bun ~/.kimi/skills/tier1380-mcp/scripts/mcp-flow.ts legacy-migration

MCP Tool Usage Examples

Shell Execute with Context

// Execute with profile and OpenClaw context
{
  "tool": "shell_execute",
  "args": {
    "command": "bun run deploy",
    "profile": "production",
    "openclaw": true,
    "workingDir": "/Users/nolarose/Projects/myapp"
  }
}

OpenClaw Operations

// Check status
{ "tool": "openclaw_status", "args": {} }

// Restart gateway
{ "tool": "openclaw_gateway_restart", "args": {} }

Profile Management

// List profiles
{ "tool": "profile_list", "args": {} }

// Switch profile
{ "tool": "profile_switch", "args": { "profile": "staging" } }

// Bind directory
{ "tool": "profile_bind", "args": { "profile": "dev" } }

Integration with Other Skills

tier1380-openclaw

# Via MCP
{ "tool": "openclaw_status" }

# Via CLI
bun run matrix:openclaw:status
ocstatus

tier1380-omega

# Via MCP shell_execute
{ "tool": "shell_execute", "args": { "command": "omega deploy staging" } }

# Via CLI
bun run omega:deploy:staging

tier1380-infra

# Via MCP shell_execute
{ "tool": "shell_execute", "args": { "command": "infra status" } }

# Via CLI
infra status

tier1380-commit-flow

# Via MCP shell_execute
{ "tool": "shell_execute", "args": { "command": "/commit '[OPENCLAW][MCP][TIER:1380] Add flow'" } }

MCP Server Development

Creating Custom MCP Servers

// Example: Custom OpenClaw MCP server
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";

const server = new McpServer({
  name: "openclaw-mcp",
  version: "1.0.0"
});

// Register tools
server.tool("openclaw_deploy", {
  environment: z.enum(["staging", "production"])
}, async ({ environment }) => {
  // Deploy logic
  return { content: [{ type: "text", text: "Deployed!" }] };
});

// Start server
const transport = new StdioServerTransport();
await server.connect(transport);

Testing MCP Servers

# Test with MCP inspector
bun run mcp:inspector

# Test specific server
bun run mcp:test bun-local

# Validate configuration
bun run mcp:validate

Environment Configuration

Required Environment Variables

# GitHub MCP
GITHUB_TOKEN=<your_token>

# OpenClaw (via Bun Secrets)
com.openclaw.gateway/gateway_token=<token>

# Bun Local MCP
BUN_RUNTIME=1
BUN_CONFIG_PREFER_OFFLINE=1

Bun Secrets Integration

# Set OpenClaw token
bun secrets set com.openclaw.gateway gateway_token <token>

# Set GitHub token
bun secrets set github token <token>

# Verify
bun secrets list

Troubleshooting

MCP Connection Issues

# Check server logs
bun run mcp:logs <name>

# Restart all servers
bun run mcp:restart-all

# Reset configuration
bun run mcp:reset

Tool Execution Failures

# Test tool directly
bun run mcp:test-tool shell_execute '{"command":"whoami"}'

# Check shell bridge
bun run mcp:shell --test

# Verify permissions
ls -la ~/.kimi/tools/unified-shell-bridge.ts

Cross-Reference Links

Related Skills

MCP Configuration Files

MCP Documentation

• MCP Specification
• Bun MCP Docs
• Claude MCP Guide

Related Skills

Tier-1380 Audit Integration

# Run audit via MCP shell_execute
{ "tool": "shell_execute", "args": { "command": "bun run tier1380:audit check README.md" } }

# Launch dashboard
{ "tool": "shell_execute", "args": { "command": "bun run tier1380:audit dashboard" } }

# Check violations via SQLite
{ "tool": "shell_execute", "args": { "command": "bun run tier1380:audit db" } }

Cross-Skill MCP Flows

# OpenClaw + Omega deployment
mcpflow profile-deploy profile=production
shell_execute "omega deploy production"

# Infrastructure + OpenClaw health
mcp:health
infra status openclaw

# Full stack with commit
mcp:status
/flow full
/commit "[PLATFORM][MCP][TIER:1380] Update flows"

Shell Aliases

alias mcpflow='bun ~/.kimi/skills/tier1380-mcp/scripts/mcp-flow.ts'
alias mcphealth='bun run mcp:health'
alias mcpmigrate='bun run mcp:migrate'

Updated: 2026-01-31 | Tier-1380 MCP Integration v1.0.0