AutoSwarm: Unify separate agents into one coordinated team that shares context, connects tools, and combines results.
What problem it solves
Individual agents have different capabilities, context, workflows, and ecosystem connections, forcing users to switch interfaces, repeat information, and coordinate subtasks manually. Users need a way to delegate a task to appropriate installed agents and consolidate their work without managing each agent separately.
Key features
Automatic agent selection — AutoSwarm selects agents according to their capabilities when it receives a task. It then connects and coordinates the selected collaborators.
Source: README
Shared context and memory — Context and memory can be shared across agents so that collaborators work from shared information rather than isolated state.
Source: README
Connected tools and workflows — The system connects complementary tools and workflows and bridges data and application ecosystems. It is available through desktop apps, command-line interfaces, IDE extensions, and skill-enabled agent environments.
Source: README
Parallel execution and result synthesis — AutoSwarm runs tasks in parallel, coordinates collaboration, and verifies and synthesizes the collected results into a coherent answer.
Source: README
Quick start
Copied verbatim from the project's README.
git clone https://github.com/HKUDS/AutoSwarm.git
cd AutoSwarm
uv tool install --force .
autoswarm setup --host codexUse AutoSwarm to read these 100 documents and summarize the key approaches and disagreements.Who it may suit
This may suit developers who already use one or more local coding agents and want to delegate broader tasks from a single conversation. It may also suit teams experimenting with multi-agent workflows across desktop, CLI, IDE, and skill-enabled environments.
Topics to discuss
Which agent-selection and task-routing rules should be configurable?
How should context sharing, approvals, and result verification work across different agents?
Which additional agent integrations, tools, and native capabilities are most valuable to support next?
Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.
Read the original README https://github.com/HKUDS/AutoSwarm#readme
ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.
Comments
…