openapi: 3.0.3
info:
title: Autonomous Software Foundry - MCP Server
description: |
A sophisticated Model Context Protocol (MCP) server implementing autonomous software generation capabilities with AI agent orchestration, self-healing systems, and comprehensive observability.
**Key Features:**
- 🤖 Multi-Agent System (Frontend, Backend, DevOps, Reviewer Agents)
- 🔄 Self-Healing Capabilities with LLM-powered failure analysis
- 🔐 Descope OAuth 2.1 + PKCE Authentication with Non-Human Identity
- 📊 Cequence AI Gateway Integration for analytics and security monitoring
- 🛠️ Full MCP Protocol Compliance (JSON-RPC 2.0)
- 🚀 Autonomous Code Generation and Deployment
**Competition Innovation:**
This MCP server demonstrates advanced autonomous capabilities by orchestrating multiple specialized AI agents to generate, test, heal, and deploy full-stack applications automatically. The self-healing loop continuously monitors generated applications and applies fixes using LLM-powered analysis.
version: 2.0.0
contact:
name: Autonomous Software Foundry
url: https://github.com/yoriichi-07/Multi_Orchestrator_MCP
license:
name: MIT
url: https://opensource.org/licenses/MIT
servers:
- url: https://server.smithery.ai/@yoriichi-07/multi_orchestrator_mcp/mcp
description: Production server on Smithery (competition hosting)
- url: http://localhost:8000
description: Local development server
security:
- DescopeBearer: []
paths:
# Health and Status Endpoints
/health:
get:
summary: Health Check
description: Check server health and basic status information
tags: [Health]
security: []
responses:
'200':
description: Server is healthy
content:
application/json:
schema:
$ref: '#/components/schemas/HealthResponse'
# MCP Protocol Core Endpoints
/mcp/initialize:
post:
summary: MCP Initialization Handshake
description: Initialize MCP client-server connection with capability negotiation
tags: [MCP Protocol]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/MCPInitializeRequest'
responses:
'200':
description: Initialization successful
content:
application/json:
schema:
$ref: '#/components/schemas/MCPInitializeResponse'
/mcp/capabilities:
get:
summary: Get MCP Server Capabilities
description: Retrieve server capabilities and supported features
tags: [MCP Protocol]
responses:
'200':
description: Server capabilities
content:
application/json:
schema:
$ref: '#/components/schemas/MCPCapabilitiesResponse'
/mcp/tools/list:
post:
summary: List Available MCP Tools
description: Get list of all available MCP tools with their schemas
tags: [MCP Tools]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/MCPListRequest'
responses:
'200':
description: List of available tools
content:
application/json:
schema:
$ref: '#/components/schemas/MCPToolsListResponse'
/mcp/tools/call:
post:
summary: Execute MCP Tool
description: Execute a specific MCP tool with provided arguments
tags: [MCP Tools]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/MCPToolCallRequest'
responses:
'200':
description: Tool execution result
content:
application/json:
schema:
$ref: '#/components/schemas/MCPToolCallResponse'
/mcp/resources/list:
post:
summary: List Available MCP Resources
description: Get list of all available MCP resources
tags: [MCP Resources]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/MCPListRequest'
responses:
'200':
description: List of available resources
content:
application/json:
schema:
$ref: '#/components/schemas/MCPResourcesListResponse'
/mcp/resources/read:
post:
summary: Read MCP Resource Content
description: Read the content of a specific MCP resource
tags: [MCP Resources]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/MCPResourceReadRequest'
responses:
'200':
description: Resource content
content:
application/json:
schema:
$ref: '#/components/schemas/MCPResourceReadResponse'
# Autonomous Software Generation Tools
/mcp/v1/tools/generate_application:
post:
summary: Generate Full-Stack Application
description: |
**Autonomous Application Generation**
This is the flagship autonomous tool that orchestrates multiple AI agents to generate a complete full-stack application from a high-level description. The system coordinates:
- Frontend Agent: React/Next.js/Vue.js generation
- Backend Agent: FastAPI/Express.js/Django generation
- DevOps Agent: Deployment and infrastructure setup
- Reviewer Agent: Code quality analysis and testing
tags: [Autonomous Generation]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/GenerateApplicationRequest'
responses:
'200':
description: Application generated successfully
content:
application/json:
schema:
$ref: '#/components/schemas/GenerateApplicationResponse'
/mcp/v1/tools/self_heal:
post:
summary: Self-Healing System
description: |
**Autonomous Self-Healing**
Advanced self-healing capability that automatically detects, analyzes, and fixes issues in generated applications using LLM-powered failure analysis. This demonstrates true autonomous operation by continuously monitoring and improving generated code.
tags: [Autonomous Healing]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/SelfHealRequest'
responses:
'200':
description: Healing process completed
content:
application/json:
schema:
$ref: '#/components/schemas/SelfHealResponse'
/mcp/v1/tools/code_review:
post:
summary: AI-Powered Code Review
description: Comprehensive code review using AI agents for quality analysis, security scanning, and improvement suggestions
tags: [Quality Assurance]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/CodeReviewRequest'
responses:
'200':
description: Code review completed
content:
application/json:
schema:
$ref: '#/components/schemas/CodeReviewResponse'
# Infrastructure and Monitoring Tools
/mcp/v1/tools/ping:
post:
summary: Connectivity Test
description: Basic connectivity and health test with scope validation
tags: [Infrastructure]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/PingRequest'
responses:
'200':
description: Ping successful
content:
application/json:
schema:
$ref: '#/components/schemas/PingResponse'
/mcp/v1/tools/system_status:
post:
summary: System Status and Metrics
description: Comprehensive system status including resource usage, agent health, and performance metrics
tags: [Infrastructure]
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/SystemStatusRequest'
responses:
'200':
description: System status retrieved
content:
application/json:
schema:
$ref: '#/components/schemas/SystemStatusResponse'
# Analytics Dashboard
/dashboard:
get:
summary: Cequence Analytics Dashboard
description: Serve the integrated Cequence AI Gateway analytics dashboard
tags: [Analytics]
responses:
'200':
description: Dashboard HTML page
content:
text/html:
schema:
type: string
components:
securitySchemes:
DescopeBearer:
type: http
scheme: bearer
bearerFormat: JWT
description: |
Descope OAuth 2.1 + PKCE JWT token with Non-Human Identity support.
**Required Scopes:**
- `tools:ping` - Basic connectivity testing
- `tools:generate` - Application generation capabilities
- `tools:review` - Code review and quality analysis
- `tools:fix` - Self-healing and fixing capabilities
- `tools:deploy` - Deployment and infrastructure management
- `admin:logs` - Access to system logs and monitoring
- `admin:config` - Configuration management
schemas:
# Health and Status
HealthResponse:
type: object
properties:
status:
type: string
example: "healthy"
service:
type: string
example: "autonomous-software-foundry"
version:
type: string
example: "2.0.0"
authentication:
type: string
example: "Descope OAuth 2.1 + PKCE"
analytics:
type: string
example: "Cequence AI Gateway"
mcp_protocol:
type: string
example: "2024-11-05"
capabilities:
type: object
properties:
tools:
type: integer
example: 12
resources:
type: integer
example: 8
# MCP Protocol Schemas
MCPInitializeRequest:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "1"
method:
type: string
example: "initialize"
params:
type: object
properties:
protocolVersion:
type: string
example: "2024-11-05"
capabilities:
type: object
clientInfo:
type: object
properties:
name:
type: string
example: "Claude Desktop"
version:
type: string
example: "1.0.0"
MCPInitializeResponse:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "1"
result:
type: object
properties:
protocolVersion:
type: string
example: "2024-11-05"
capabilities:
$ref: '#/components/schemas/MCPCapabilities'
serverInfo:
$ref: '#/components/schemas/MCPServerInfo'
MCPCapabilities:
type: object
properties:
experimental:
type: object
logging:
type: object
prompts:
type: object
properties:
listChanged:
type: boolean
example: true
resources:
type: object
properties:
subscribe:
type: boolean
example: true
listChanged:
type: boolean
example: true
tools:
type: object
properties:
listChanged:
type: boolean
example: true
MCPServerInfo:
type: object
properties:
name:
type: string
example: "autonomous-software-foundry"
version:
type: string
example: "2.0.0"
description:
type: string
example: "MCP server with AI agent orchestration, self-healing capabilities, and Cequence AI Gateway integration"
MCPCapabilitiesResponse:
type: object
properties:
apiVersion:
type: string
example: "2024-11-05"
capabilities:
$ref: '#/components/schemas/MCPCapabilities'
serverInfo:
$ref: '#/components/schemas/MCPServerInfo'
MCPListRequest:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "2"
method:
type: string
example: "tools/list"
MCPToolsListResponse:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "2"
result:
type: object
properties:
tools:
type: array
items:
$ref: '#/components/schemas/MCPTool'
MCPTool:
type: object
properties:
name:
type: string
example: "generate_application"
description:
type: string
example: "Generate a complete full-stack application using AI agent orchestration"
inputSchema:
type: object
properties:
type:
type: string
example: "object"
properties:
type: object
required:
type: array
items:
type: string
MCPToolCallRequest:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "3"
method:
type: string
example: "tools/call"
params:
type: object
properties:
name:
type: string
example: "generate_application"
arguments:
type: object
MCPToolCallResponse:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "3"
result:
type: object
properties:
content:
type: array
items:
type: object
properties:
type:
type: string
example: "text"
text:
type: string
# Tool-specific Request/Response Schemas
GenerateApplicationRequest:
type: object
required:
- project_description
properties:
project_description:
type: string
description: High-level description of the application to generate
example: "Create a task management app with user authentication, real-time updates, and mobile-responsive design"
project_type:
type: string
enum: [fullstack, frontend, backend, mobile]
default: fullstack
description: Type of application to generate
framework_preferences:
type: object
properties:
frontend:
type: string
enum: [react, nextjs, vue, angular]
example: "nextjs"
backend:
type: string
enum: [fastapi, express, django, flask]
example: "fastapi"
database:
type: string
enum: [postgresql, mysql, mongodb, sqlite]
example: "postgresql"
additional_requirements:
type: array
items:
type: string
example: ["user authentication", "real-time chat", "file uploads"]
GenerateApplicationResponse:
type: object
properties:
status:
type: string
example: "success"
project_id:
type: string
example: "proj_12345"
generated_files:
type: array
items:
type: object
properties:
path:
type: string
example: "src/components/TaskList.tsx"
content:
type: string
agent:
type: string
example: "frontend"
agent_reports:
type: object
properties:
frontend:
$ref: '#/components/schemas/AgentReport'
backend:
$ref: '#/components/schemas/AgentReport'
devops:
$ref: '#/components/schemas/AgentReport'
reviewer:
$ref: '#/components/schemas/AgentReport'
deployment_info:
type: object
properties:
url:
type: string
example: "https://task-app-12345.smithery.ai"
status:
type: string
example: "deployed"
AgentReport:
type: object
properties:
agent_type:
type: string
example: "frontend"
status:
type: string
enum: [completed, in_progress, failed]
files_generated:
type: integer
example: 8
technologies_used:
type: array
items:
type: string
example: ["React", "TypeScript", "Tailwind CSS"]
completion_time:
type: string
format: date-time
notes:
type: string
example: "Generated responsive UI with modern design patterns"
SelfHealRequest:
type: object
required:
- project_id
- issue_description
properties:
project_id:
type: string
description: ID of the project to heal
example: "proj_12345"
issue_description:
type: string
description: Description of the issue to fix
example: "API endpoints are returning 500 errors on user authentication"
severity:
type: string
enum: [low, medium, high, critical]
default: medium
auto_deploy:
type: boolean
default: false
description: Whether to automatically deploy fixes
SelfHealResponse:
type: object
properties:
status:
type: string
example: "healing_completed"
healing_id:
type: string
example: "heal_67890"
issues_found:
type: array
items:
type: object
properties:
type:
type: string
example: "authentication_error"
severity:
type: string
example: "high"
description:
type: string
example: "JWT token validation failing"
file:
type: string
example: "src/auth/middleware.py"
fixes_applied:
type: array
items:
type: object
properties:
issue_type:
type: string
example: "authentication_error"
fix_description:
type: string
example: "Updated JWT token validation logic"
files_modified:
type: array
items:
type: string
example: ["src/auth/middleware.py", "tests/test_auth.py"]
test_results:
type: object
properties:
passed:
type: integer
example: 47
failed:
type: integer
example: 0
coverage:
type: number
format: float
example: 94.5
CodeReviewRequest:
type: object
required:
- project_id
properties:
project_id:
type: string
example: "proj_12345"
review_type:
type: string
enum: [security, performance, quality, full]
default: full
include_suggestions:
type: boolean
default: true
CodeReviewResponse:
type: object
properties:
review_id:
type: string
example: "review_98765"
overall_score:
type: number
format: float
example: 8.7
categories:
type: object
properties:
security:
type: number
format: float
example: 9.2
performance:
type: number
format: float
example: 8.5
maintainability:
type: number
format: float
example: 8.9
test_coverage:
type: number
format: float
example: 94.5
issues:
type: array
items:
type: object
properties:
severity:
type: string
enum: [info, warning, error, critical]
category:
type: string
example: "security"
message:
type: string
example: "Potential SQL injection vulnerability"
file:
type: string
example: "src/database/queries.py"
line:
type: integer
example: 45
suggestion:
type: string
example: "Use parameterized queries instead of string concatenation"
PingRequest:
type: object
properties:
message:
type: string
example: "Hello from MCP client"
PingResponse:
type: object
properties:
status:
type: string
example: "pong"
timestamp:
type: string
format: date-time
server_info:
type: object
properties:
python_version:
type: string
example: "3.11.5"
platform:
type: string
example: "Linux-5.4.0-x86_64"
cpu_count:
type: integer
example: 4
memory_gb:
type: number
format: float
example: 16.0
authentication:
type: object
properties:
user_id:
type: string
example: "machine_user_123"
is_machine:
type: boolean
example: true
scopes:
type: array
items:
type: string
example: ["tools:ping", "tools:generate"]
SystemStatusRequest:
type: object
properties:
include_metrics:
type: boolean
default: true
SystemStatusResponse:
type: object
properties:
status:
type: string
example: "healthy"
timestamp:
type: string
format: date-time
system_metrics:
type: object
properties:
cpu:
type: object
properties:
usage_percent:
type: number
format: float
example: 23.5
count:
type: integer
example: 4
memory:
type: object
properties:
total_gb:
type: number
format: float
example: 16.0
used_gb:
type: number
format: float
example: 8.2
usage_percent:
type: number
format: float
example: 51.25
disk:
type: object
properties:
total_gb:
type: number
format: float
example: 100.0
free_gb:
type: number
format: float
example: 65.3
usage_percent:
type: number
format: float
example: 34.7
agent_status:
type: object
properties:
frontend_agent:
$ref: '#/components/schemas/AgentStatus'
backend_agent:
$ref: '#/components/schemas/AgentStatus'
devops_agent:
$ref: '#/components/schemas/AgentStatus'
reviewer_agent:
$ref: '#/components/schemas/AgentStatus'
active_projects:
type: integer
example: 3
healing_queue:
type: integer
example: 1
AgentStatus:
type: object
properties:
status:
type: string
enum: [idle, busy, error]
example: "idle"
last_activity:
type: string
format: date-time
current_task:
type: string
example: "Generating React components"
tasks_completed:
type: integer
example: 47
MCPResourcesListResponse:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "4"
result:
type: object
properties:
resources:
type: array
items:
$ref: '#/components/schemas/MCPResource'
MCPResource:
type: object
properties:
uri:
type: string
example: "projects://active/list"
name:
type: string
example: "Active Projects"
description:
type: string
example: "List of currently active projects in the system"
mimeType:
type: string
example: "application/json"
MCPResourceReadRequest:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "5"
method:
type: string
example: "resources/read"
params:
type: object
properties:
uri:
type: string
example: "projects://active/list"
MCPResourceReadResponse:
type: object
properties:
jsonrpc:
type: string
example: "2.0"
id:
type: string
example: "5"
result:
type: object
properties:
contents:
type: array
items:
type: object
properties:
uri:
type: string
example: "projects://active/list"
mimeType:
type: string
example: "application/json"
text:
type: string
description: The resource content
tags:
- name: Health
description: Health check and server status endpoints
- name: MCP Protocol
description: Core MCP protocol implementation (JSON-RPC 2.0)
- name: MCP Tools
description: MCP tool listing and execution
- name: MCP Resources
description: MCP resource management
- name: Autonomous Generation
description: AI-powered autonomous application generation
- name: Autonomous Healing
description: Self-healing and automated fixing capabilities
- name: Quality Assurance
description: Code review and quality analysis tools
- name: Infrastructure
description: System monitoring and infrastructure tools
- name: Analytics
description: Cequence AI Gateway analytics and monitoring
