guilds.run

Run inspectable AI agent teams with shared memory and workflows

Website

Meet guilds.run

guilds.run is an open-source system for operating teams of AI agents on a local machine. Users organize agents into groups with roles, routines, and shared work, then inspect the records behind each run. Those records include the prompt, tool calls and results, memory changes, and the agent responsible for a note.

Shared tables, scripts, notes, and memory help work continue between runs. The Studio feature can review a team's behavior and propose changes to instructions or configuration, with the user reviewing and publishing those updates. Recipes and connectors provide starting points for workflows such as research, growth, or support. The system supports different model choices, while external services and model usage depend on the user's connections.

Discussion

Tried it? Share a useful thought or ask a question.

Product details

Website
guilds.run
Status
Launched
Listed

Explore related topics