Civic Labs
Here you'll find a suite of experiments and tools, some just getting started, some ready to test, all open to feedback and community input.
Each project is presented as a flask, representing different stages of experimental development. Learn more about flask status →.
Flasks are standalone services, libraries or tools, that can be used independently or together to build more complex applications.
Explore the flasks below or get in touch if you want to contribute, try things early, or shape what comes next.
🚀 Getting Started
Want to try out these experiments? Get started here.
💬 Feedback & Contribution
We're building in the open and love community input. Learn how to contribute.
🧪 Flasks
Our current experiments focus on Model Context Protocol (MCP) tools and AI security. These tools help developers build safer, more controlled AI applications with proper identity and authorization.
A hosted MCP Manager unifying and orchstrating multiple MCP servers, focusing on auth and security
Wrap any MCP server in a configurable and flexible security layer
LLM-based threat detection for prompts and tool calls
Middleware hook system for MCP servers that powers guardrails and more
AI assistant for the optimisation of internal operations and processes
📚 Concepts & Architecture
Understanding the building blocks behind our experiments.
What is MCP and why it matters for AI applications
Guardrails as a protection layer
Understanding prompt injection attacks & LLM safety
OAuth2, granular permissions, and consent
A middleware layer around MCP APIs
Retrieval strategies for LLMs