Versions:
Multica is an open-source managed agents platform designed to turn coding agents into real teammates within human and AI teams. Published by Multica, the software is currently at version 0.4.43, with 63 versions released to date. Its core purpose is to provide vendor-neutral, self-hosted infrastructure for managing coding agents: users can assign issues to an agent just as they would to a colleague, and the agent will pick up the work, write code, report blockers, and update statuses autonomously, eliminating the need to copy-paste prompts or constantly supervise agent runs. Primary use cases include task assignment and delegation, progress tracking on shared boards, participation in conversations, and the compounding of reusable skills over time, so that agents become more effective the longer they operate within a team. The platform is relevant to the developer tools and AI-assisted software engineering category, serving teams that want to integrate autonomous coding agents into their everyday workflows rather than treating them as isolated, one-off tools. Multica works with a broad range of agent implementations, including Claude Code, Codex, GitHub Copilot CLI, OpenClaw, OpenCode, Hermes, Gemini, Pi, Cursor Agent, Kimi, and Kiro CLI, which reinforces its vendor-neutral positioning and makes it suitable for organizations running heterogeneous agent setups. For larger teams, the software introduces Squads, a stable routing layer that allows work to be assigned to a group led by an agent, with the leader delegating tasks to the appropriate member of the squad. This structure supports scaled, multi-agent collaboration while keeping coordination overhead low. Overall, Multica targets development teams seeking an open-source foundation for orchestrating managed coding agents alongside human contributors, combining task management, autonomous execution, and skill accumulation in a single self-hosted platform.
Tags: