clqn

mcp
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 5 GitHub stars
Code Gecti
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested

Bu listing icin henuz AI raporu yok.

SUMMARY

CLI-first cloud workspaces for large teams and their agents

README.md
clqn

clqn.org · Download · Documentation

Build status Node.js 22 or newer npm version MIT license

Why clqn?

Collaborating on long-horizon development tasks can get complicated quickly. Bloated context, merge conflicts, and unclear task ownership are common, especially when each collaborator works with several agents.

Existing solutions introduce a lot of setup overhead. clqn is an extra lightweight CLI for people and their own AI agents to easily coordinate work.

clqn stands for clan. Each clan has a leader who invites members and registers an orchestrator agent. Members (and the leader) register worker agents.

Installation

Install with npm (Node.js 22+):

npm install -g clqn-cli
clqn --help

Alternatively, install with Homebrew:

brew tap rli8145/clqn https://github.com/rli8145/clqn
brew install clqn

Usage

1. Sign in and create a clan

Use the API URL from your clan operator.

export CLQN_API_URL="<operator-api-url>"
clqn login --github
clqn init "My project"
clqn invite [email protected]

Invited? Sign in and run clqn join <invite-code>.

2. Connect your agents

The leader registers the orchestrator (optional but recommended). Otherwise, the leader supervises worker agents manually. A strong reasoning model is recommended for the orchestrator. Each member registers their own workers.

clqn agent add Coordinator --orchestrator
clqn agent add codex

Connect each agent to MCP using its returned token: agent setup.

Set token limits, allowed models, and approval preferences for your agents:

clqn limits --tokens 100000 --models provider/model --control approve

3. Share context and coordinate work

printf 'Run npm test before committing.\n' | clqn context write conventions --expected-revision 0
clqn context read conventions

For code tasks, run from your project clone with Git and an authenticated GitHub CLI (gh). Each agent uses its own credentials.

# Orchestrator
clqn task add login-fix --code
clqn task assign login-fix --agent codex

# Worker
clqn task start login-fix
cd <printed-worktree-path>
# Make changes, test, and commit.
clqn task submit login-fix

# Orchestrator
clqn task check login-fix
clqn task merge login-fix

# Worker
clqn task done login-fix

As the leader, run clqn repo link once to set up the GitHub webhook. While MCP is connected, changes to origin/main automatically update clean local main checkouts.

Use --alias if you want a longer agent name or task description with a separate short name. See custom aliases.

4. Follow progress

clqn status
clqn log

Use clqn --help to explore all commands. Add --json for scripting.

Documentation

Contributing

Report an issue or read the contribution guide before submitting a pull request.

Yorumlar (0)

Sonuc bulunamadi