systemprompt-template

Production AI agent mesh in 3 commands. MCP servers, playbooks, and multi-agent orchestration built on systemprompt-core.

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

View docs

Command

claude mcp add --transport stdio systempromptio-systemprompt-template docker run -i systempromptio/systemprompt-template \
  --env DATABASE_URL="PostgreSQL connection string" \
  --env SYSTEMPROMPT_API_KEY="your-api-key (optional)"

How to use

systemprompt-template provides a production-grade AI agent mesh built on the systemprompt platform. It hosts multiple MCP services (3 included in the template) that coordinate agents, tools, and playbooks to enable autonomous workflows and orchestration. The template emphasizes an end-to-end AI evolution model—from single-turn chat interactions to tool usage, autonomous agents, and mesh-level orchestration. You’ll use the provided CLI commands to login, create a tenant, start the services, and then interact with the hosted agents and tools via the dashboard and predefined playbooks. Typical workflows include running tool-enabled tasks, deploying content or pipelines, and leveraging the four levels of capability to compose complex agent-based behaviors. The Playbooks section shows ready-made, machine-readable instructions that ensure deterministic behavior for common operations, reducing hallucinations and miscommands in AI-driven workflows.

How to install

Prerequisites

Docker (for local PostgreSQL and runtime) or a native Rust build environment if you prefer building from source
PostgreSQL (local or remote) as the template relies on a database
Optional: systemprompt CLI tools if you want direct CLI control in addition to the MCP interfaces

Installation steps (using Docker)

Install Docker on your machine according to the official docs for your OS.
Ensure PostgreSQL is accessible (local or remote) and note the connection URL.
Run the MCP server container: docker run -i --env DATABASE_URL="postgres://user:password@localhost:5432/dbname" --env SYSTEMPROMPT_API_KEY="your-api-key" systempromptio/systemprompt-template
Open the MCP dashboard at http://localhost:8080 (or the port you expose) to manage agents, tools, and playbooks.

Alternative: Build from source

Install Rust 1.75+ and just (cargo install just).
Clone the repository and follow the project’s build instructions (usually just: just build).
Run the built binary locally, ensuring you connect to PostgreSQL with the appropriate environment variables.

Additional notes

Environment variables and configuration:

DATABASE_URL should point to your PostgreSQL instance. The template expects a usable database with the necessary schema.
SYSTEMPROMPT_API_KEY (optional) can enable cloud features or analytics if supported by your deployment.
If running with Docker, ensure the container can reach PostgreSQL (network access, proper host/port). Common issues:
Database connection failures: verify credentials, network access, and that the database is up and migrated.
Port conflicts: ensure the dashboard port (default 8080) is available or adjust your container run parameters.
Authentication: use the provided login/tenant commands to authenticate and create your tenant before starting workflows. Configuration options:
Adjust MCP service count, timeouts, and playbook selections via the dashboard or CLI, depending on how you deploy (Docker image exposed configuration options may exist in environment variables or mounted config files).