EVALUATE FOR FREE.

systemprompt.io is self-hosted AI governance infrastructure. It is a single Rust binary (under 50MB) that governs every AI tool call your agents make. It is not a SaaS product. It is not a framework. It is infrastructure you own, run on your servers, and extend with your own code.

Every tool call passes through four enforcement layers before execution: scope-based access control, secret detection, rate limiting, and an immutable audit trail. The governance evaluation completes in under 5 milliseconds. No perceptible latency. No performance tax on your AI workflow.

PostgreSQL is the only external dependency. The binary handles HTTP serving, MCP governance, agent orchestration, content management, analytics, and background jobs. What others assemble from six services, this ships as one.

The template repository is the fastest path from zero to governed AI agents. It includes a pre-configured binary, sample governance rules, and everything you need to see the enforcement pipeline in action.

Prerequisites

You will need Git, the Rust toolchain (rustup.rs), and PostgreSQL 15+ running locally or accessible over the network. The build compiles to a single binary with no runtime dependencies beyond PostgreSQL.

Clone and Build

git clone https://github.com/systempromptio/systemprompt-template.git
cd systemprompt-template
cargo build --release

Configure

Copy the example environment file and set your database connection string. The template ships with sensible defaults. You only need to configure PostgreSQL credentials and your AI provider API key.

cp .env.example .env
# Edit .env with your PostgreSQL connection string
# Add your Anthropic API key (or other provider)

Run

./target/release/systemprompt serve

The server starts on port 3000 by default. Hit the health endpoint to verify everything is running:

curl http://localhost:3000/health

You should see a JSON response confirming the server is ready and the database is connected. From here, every AI tool call through this server is governed.

The template ships with pre-configured governance rules that demonstrate every enforcement layer. Here is what to look for during your evaluation and how to test each capability.

Governance Pipeline

Connect an AI agent and make a tool call. Watch the governance pipeline evaluate scope, scan for secrets, enforce rate limits, and record the decision, all in under 5 milliseconds. Try calling the same tool with different agent scopes: admin-scope agents pass, user-scope agents are denied. The enforcement is immediate and deterministic.

Secret Scanning

Inject a test credential into a tool call input: an AWS access key, a GitHub token, or a PEM private key. The secret scanner blocks the call before it reaches any external service. The credential never leaves the governance layer. Check the audit trail to see exactly which pattern matched and why the call was denied.

Audit Trail

Every governance decision is recorded with the full evaluation context: which agent, which tool, which rules fired, what policy matched, and why. Query the audit trail through the CLI or API. Export it to your SIEM. This is the evidence your security team reviews during an audit.

Cost Attribution

Every AI request is attributed to its agent with token counts, model, latency, and cost. Run multiple agents and check the per-agent cost breakdown. No hidden spend. No shared buckets. Every token accounted for.

The template is not a toy example. It is a production-ready starting point that compiles into your own binary. Everything inside is yours to own, modify, and extend.

Configuration Layer

All governance rules, agent definitions, and server settings are YAML files checked into Git. No database configuration. No admin panels. Your infrastructure-as-code workflow applies directly. Review governance changes in pull requests. Roll back with git revert.

Pre-Configured Agents

The template ships with sample agents at different scope levels: admin, user, and restricted. Each agent has different tool access permissions so you can immediately test the governance pipeline's scope enforcement without writing any configuration from scratch.

MCP Server Registry

MCP servers are registered, authenticated, and governed. The template includes example server registrations with per-server OAuth2 configuration. Every tool call through every MCP server passes through the same governance pipeline.

Extension Points

systemprompt.io is a library, not a framework. Add custom API routes, database tables, background jobs, and page providers through Rust extension traits. Your code compiles into your binary. No plugin runtime. No performance overhead. No version conflicts.

The evaluation is just the beginning. Once you have seen the governance pipeline in action, there are several paths forward depending on your needs.

Self-Guided

The documentation covers deployment, configuration, extensions, and every service in the binary. The guides walk through specific workflows like connecting MCP servers, configuring compliance rules, and setting up analytics exports. The live demo shows four governance layers executing in real time in under six minutes.

Talk to the Founder

For teams evaluating enterprise deployment, custom integrations, or commercial licensing, book a 30-minute technical call below. Discuss your architecture, compliance requirements, and deployment timeline directly with the person who built it.