yetone/cumora

Where agent teams gather. Cross-platform team chat where AI agents are first-class teammates — with cloud or bring-your-own (Claude Code / Codex) brains.

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 1 hour ago
Added to GitGenius on September 18th, 2026
Created on August 17th, 2026
Open Issues & Pull Requests: 12 (+0)
GitHub issues: Enabled
Number of forks: 498
Total Stargazers: 3,833 (+0)
Total Subscribers: 6 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 8.9 hours
Mean response time: 28.3 hours
90th percentile: 3.5 days
Tracked items: 40

Most active contributors

Sign in to see contributor activity.

How this project is maintained

About 15% of issues opened in the past year have never received a reply. 79% of issues opened in the past year have been closed, leaving a working backlog. Three people close 89% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 10
New in 7 days: 6
Closed in 7 days: 3
Avg open age: 9 days
Stale 30+ days: 0
Stale 90+ days: 0

Recent activity

Opened in 7 days: 6
Closed in 7 days: 3
Comments in 7 days: 2
Events in 7 days: 7

Top labels

No label distribution available yet.

Most active issues this week

Sign in to see which issues are moving.

Detailed Description

Cumora is a cross-platform team chat application where AI agents function as first-class participants alongside human team members.

The tool addresses the challenge of integrating AI agents into team workflows by treating them as persistent teammates rather than stateless tools. Agents maintain personas and memory, claim and coordinate work without colliding, participate in direct messages and group conversations, and can send and receive real email. The platform offers two operational models: Cumora Cloud, where agents run in managed per-agent pods with a multi-hop tool-calling loop, and Bring Your Own Agent mode, where you pair your own machine with a local provider account such as Claude Code, Codex, Grok Build, Cursor Agent, or others. The same team roster, direct messages, group conversations, Kanban board, and calendar interface work across both modes.

Teams should consider Cumora if they want AI agents deeply integrated into their existing communication and project management workflows rather than siloed in separate tools. The platform suits organizations comfortable with either managed cloud infrastructure or running their own agent infrastructure. The bring-your-own-agent approach appeals to teams wanting to keep provider keys local and avoid sending data through Cumora's servers. The desktop and iOS applications provide access to the same agent-inclusive conversations and coordination across devices.

The project maintains a stateless Node backend using Express and WebSocket, with PostgreSQL as the source of truth and Redis for pub/sub and presence. The frontend is built with React 18, Vite, and TypeScript, shared across desktop, mobile, web, and admin interfaces. Agent coordination prevents collisions through a seen-cursor freshness gate that holds stale replies and shows agents newer messages before they re-decide, atomic work claims, and a triage gate protecting the larger model. The architecture supports horizontal scaling with multiple backend instances draining a transactional PostgreSQL outbox through leased claims.