Developer Resources
Build with Calliope AI. Govern anything you ship.
Documentation, quickstart guides, a full API reference, and the Zentinelle SDK to add governance to any agent you build. Everything you need to get started — and to scale into production inside your own perimeter.
Get started
Resources for every stage of your journey.
Whether you're just getting started or scaling to production, we have resources to help — documentation, tutorials, community support, and direct access to the team.
Quickstart
Getting started, the fastest path.
New to Calliope AI? Four steps from zero to building.
- 1. Sign up — pick a plan at calliope.ai/pricing. A free tier is available to start.
- 2. Read the quickstart — the fastest orientation lives in the docs at docs.calliope.ai.
- 3. Join Discord — ask questions and connect with the community in the Calliope AI Discord.
- 4. Explore the examples — sample code and integrations live in our GitHub repositories.
Documentation
Guides for the whole Workbench.
From first steps to advanced governance configuration. Visit docs.calliope.ai for the complete documentation.
Platform Overview
Understand The Workbench and how its components fit together inside your perimeter.
AI Lab
Set up notebooks, connect data sources, and build pipelines with AI assistance.
AI IDE
Configure Completions, Chat, and Agent modes for your development workflow.
Chat Studio
Register datasets and build chat interfaces over your own data.
Deep Agent
Build RAG pipelines and research workflows with the Deep Data Agent.
Langflow
Visual workflow builder tutorials for assembling agents and chains.
Governance
Policy configuration, audit logging, and compliance mapping with Zentinelle.
Tutorials
Step-by-step guides for the most common build-and-ship use cases.
API Reference
Programmatic access to every service.
Complete API documentation for Calliope AI services, available at docs.calliope.ai/api.
- REST API endpoints — call platform services directly from your own code.
- Authentication & authorization — token-based access scoped to your Workbench session.
- Rate limits & quotas — documented limits and budget controls per service.
- SDK libraries — official clients for Python, TypeScript, and Go.
Zentinelle SDK
Govern any AI agent — not just ours.
The Zentinelle SDK adds governance to any agent you build — policy enforcement, audit logging, and compliance, dropped into your own code. It's MIT-licensed and public on GitHub.
- Policy enforcement hooks — gate tool calls and data access in real time.
- Audit logging integration — every prompt, response, and tool call attributed and queryable.
- Compliance controls — map agent behavior to the frameworks you answer to.
- Rate limiting & budget caps — keep spend and blast radius bounded.
- Content scanning — catch sensitive data and malicious prompts before they ship.
Available in Python, TypeScript, and Go — and it works with the agent frameworks your team already uses.
Open Source
Built in the open.
Open-core: the Astrolift and Zentinelle core, plus the SDK and CLI, are MIT-licensed and public — newer enterprise capabilities and the Workbench are commercial. We open-source a path to enterprise. Find us on github.com/calliopeai.
Astrolift
The Kubernetes runtime that hosts the Workbench in your own account — AWS today, built to port. Open-core (MIT core).
View on GitHub →Zentinelle
Agent GRC — the governance layer for auditable, policy-enforced AI. Open-core (MIT core).
View on GitHub →Zentinelle SDK
Official client libraries in Python, TypeScript, and Go for adding governance to any agent.
View on GitHub →Langflow
We contribute to the visual AI workflow builder, provisioned as an add-on in the Workbench.
View on GitHub →Example integrations
Sample code for common patterns — connecting data, wiring agents, enforcing policy.
View on GitHub →Issue tracking
Report bugs, request features, and follow development across our public repositories.
View on GitHub →Community & Support
Builders helping builders.
The Calliope AI Discord is where builders connect — and enterprise teams get a direct line to engineering.
- #help — get answers to technical questions from the community and team.
- #showcase — share what you're building.
- #feedback — feature requests and suggestions.
- #announcements — product updates and news.
Enterprise support adds dedicated channels (private Slack or Teams), SLAs with guaranteed response times, architecture review with engineering, and custom onboarding and training for your team. Contact us to learn more.
Questions? Ideas?
Let's talk.
We're builders too. If you have questions, feedback, or ideas for the platform, we want to hear from you.
