mindX API

mindX provides a production-ready API for agents, UIs, and external systems with enterprise security, authentication, and monitoring. Read the docs at the interactive Swagger UI when the backend is running.

Port: The mindX backend runs by default on port 8000. In the UI, the API tab (main menu) includes a direct link to the API information mindX provides: the interactive Swagger UI at that base URL.

📚 Complete Documentation: See API Documentation for comprehensive API reference with authentication, endpoints, examples, and SDK usage patterns.

---

🔐 Production API Features

Enterprise Security

• Wallet-Based Authentication: Ethereum signature-based authentication with session tokens
• Advanced Rate Limiting: Multi-algorithm rate limiting with client reputation tracking
• CORS Protection: Strict CORS configuration for production security
• Input Validation: Comprehensive request validation and sanitization
• Security Middleware: Real-time threat detection and monitoring

Production Endpoints

• Health Monitoring: /health and /health/detailed for system monitoring
• Agent Management: /agents for agent registry and lifecycle management
• Authentication: /users/register-with-signature and session validation
• Vault Operations: Encrypted storage for user data and API keys

---

Base URL and interactive docs

• Base URL (default): http://localhost:8000
• Production Base URL: https://agenticplace.pythai.net (when deployed)
• Interactive API docs (Swagger UI): http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc (if enabled)

Use http://localhost:8000/docs to browse all endpoints, try requests, and inspect request/response schemas.

📖 Complete API Reference: API Documentation - Comprehensive guide with authentication flows, endpoint reference, examples, and SDK usage

---

Main route groups

Prefix	Description
/agenticplace	AgenticPlace integration (agent calls, Ollama ingest, CEO status)
/api	LLM provider APIs, admin Ollama, provider registry, monitoring
/api/monitoring/inbound	Inbound request metrics (latency ms, bytes, req/min); see docs/monitoring_rate_control.md
/mindxagent	mindXagent interaction and memory logging
/rage	RAGE routes (if enabled)
/llm	LLM routes (keys, test, generate)

Monitoring and rate control apply in both directions (inbound and outbound). See docs/monitoring_rate_control.md for scientific network and data metrics (ms, bytes, req/min, tokens).

---

AgenticPlace (mindX as provider)

AgenticPlace uses mindX as its provider: the frontend calls the mindX backend; mindX uses Ollama (and other LLM providers) for inference.

Endpoints:

Method	Endpoint	Description
POST	/agenticplace/agent/call	Call mindXagent through specified agent (CEO, mastermind, mindx, suntsu, pythai)
POST	/agenticplace/ollama/ingest	Ingest prompt through Ollama AI; optionally store in memory
GET	/agenticplace/ceo/status	Get CEO status and seven soldiers

AgenticPlace frontend config:

• Set VITE_MINDX_API_URL to the mindX backend base URL (e.g. http://localhost:8000). Default is http://localhost:8000.
• AgenticPlace then sends requests to ${baseUrl}/agenticplace/agent/call and ${baseUrl}/agenticplace/ollama/ingest.

---

Connecting mindX to Ollama

For AgenticPlace to use Ollama via mindX:

1. Run Ollama (e.g. on the same machine or reachable host).

- Default port: 11434. - Test: curl http://localhost:11434/api/tags

1. Configure mindX so the backend uses that Ollama instance:

- Primary: models/ollama.yaml → base_url: http://localhost:11434 (or your Ollama URL). - Override: MINDX_LLM__OLLAMA__BASE_URL in .env, or llm.ollama.base_url in data/config. - See llm/RESILIENCE.md and docs/rate_limiting_optimization.md for fallback and rate limits.

1. Start the mindX backend (e.g. port 8000).

- AgenticPlace calls mindX; mindX calls Ollama for /agenticplace/ollama/ingest and for agent inference when using Ollama.

1. Verify:

- Open http://localhost:8000/docs and try GET /api/admin/ollama/models (or the equivalent admin Ollama endpoint). - Try POST /agenticplace/ollama/ingest with body {"prompt": "Hello", "model": "your-model"}.

---

Monitoring and rate control (both directions)

Whether mindX is ingesting, providing inference, or services, monitoring and rate control are essential in both directions. Actual network and data metrics (scientific units: ms, bytes, req/min) are defined in docs/monitoring_rate_control.md. Inbound: GET /api/monitoring/inbound. Outbound: rate limiter and provider metrics.

Related docs

• AgenticPlace: docs/AgenticPlace_Deep_Dive.md (API reference section).
• Ollama: docs/ollama_api_integration.md, api/ollama/ollamaapi.md, models/ollama.yaml.
• LLM factory: llm/llm_factory.md.
• Resilience / fallback: llm/RESILIENCE.md.
• Monitoring / rate control: docs/monitoring_rate_control.md, docs/rate_limiting_optimization.md.

---

---