Manage coding agents as team members with task queues, skill libraries, and runtime monitoring for AI teams.

Multica is an open-source, self-hosted web app and server that manages coding agents as integrated team members. It provides the infrastructure necessary to assign tasks, track progress, and monitor the execution of autonomous agents within a unified workspace. The software is deployed as a server with a supporting local daemon that connects the web interface to the actual compute environments where agents execute their work.
Users interact with the platform via a web application and a command line interface. The system allows teams to treat AI agents as first-class colleagues who can pick up issues, write code, and report blockers proactively. It supports a wide range of agent providers, making it a vendor-neutral layer for AI-assisted development. This approach removes the need for manual prompt copy-pasting and constant supervision of agent runs.
The architecture consists of a Next.js frontend, a Go backend using the Chi router, and a PostgreSQL 17 database with pgvector. It utilizes a local daemon to auto-detect available agent CLIs on the host machine's path, which then reports capabilities back to the server. This allows the system to know exactly where to route work based on the available runtimes. The platform is designed for teams that want to scale their output by multiplexing human engineers with a fleet of autonomous agents, allowing a small team to operate with the capacity of a much larger organization.
Multica is infrastructure for human and AI hybrid teams, focusing on the management and orchestration of coding agents.
An AI coding agent that integrates with terminals and IDEs while supporting over 75 LLM providers.
An agentic workspace for generating webHML prototypes, pitch decks, and motion graphics using local-first AI agentsL agents.
Join our newsletter to get shiny new open source software delivered to your inbox. Unsubscribe anytime.