Deployment

HTS MCP can be deployed directly or via Docker. All three interfaces (MCP, REST, CLI) share the same codebase and configuration.

Deployment Options

Method	Best For	Notes
Direct	Development, MCP usage	Run with `uv run`
Docker	Production REST API	Includes PostgreSQL + pgvector

Quick Start

Direct (Development)


git clone https://github.com/mfbaig35r/hts-api.git
cd hts-api
uv sync
cp .env.example .env
# Edit .env with your DATABASE_URL and OPENAI_API_KEY
 
# MCP Server
uv run python -m hts.mcp
 
# REST API
uv run uvicorn hts.api:app --host 0.0.0.0 --port 8000
 
# CLI
uv run hts status

Docker


docker compose up -d

See Docker for full setup details.

Requirements

Requirement	Version	Purpose
Python	3.11+	Runtime
PostgreSQL	14+	Database
pgvector	0.5+	Vector similarity search
OpenAI API	-	Embeddings + LLM enrichment

Environment

All configuration is via environment variables. See Configuration for the complete reference.

Required:

DATABASE_URL: PostgreSQL connection string
OPENAI_API_KEY: OpenAI API key

Optional:

API_KEY: REST API authentication
RATE_LIMIT_ENABLED: Enable rate limiting