Kanban-Style AI Coding Agent Orchestration | cadre.rocks

Run coding agents like a team. Keep control of what ships.<br>Turn well-defined tasks into parallel agent runs. Route work to the right model, answer blockers, review plans and diffs, and approve every outcome from one workspace.<br>Choose your download Watch the workflow<br>Free local boards · No credit card · Windows, macOS & Linux

ACTIVE3 agents running

AWAITING1 waiting on you

REVIEW2 ready to inspect

COTP-142Harden command approvals<br>Payments · Codex running<br>RUNNING<br>CCTP-151Refactor billing service<br>Payments · Claude Code running<br>RUNNING

CCTP-166Choose retry policy<br>Needs input

The whole agent workflow, in one place.<br>Follow a task from brief to branch, blockers, plan, diff and approval—then take the same work across devices when your team grows.

01 / 09<br>Give agents a real brief.<br>Create tasks with descriptions, acceptance criteria, images and tags—then move them through the workflow your team already understands.<br>Custom sections and columns<br>Agent, tag and role routing<br>Context stays attached to the task<br>← Ⅱ Pause →<br>Showing Payments: Give agents a real brief.

AAcme Team plan<br>New board<br>LRLucas Rodrigues [email protected]<br>Filter by title, code or description… Type: all Priority: all Agent: all<br>Manual<br>Add invoice export<br>TP-171 Medium<br>frontend<br>No agent<br>3h

Migrate webhook retries<br>TP-168 Low<br>backend<br>No agent

Claude Code, Codex<br>Harden command approvals<br>TP-142 Urgent<br>securitybackend<br>COCodex

Refactor billing service<br>TP-151 High<br>backend<br>CCClaude Code

Manual<br>Choose retry policy<br>TP-166 High<br>backend<br>CCClaude Code

Codex<br>Add repository tests<br>TP-158 Medium<br>tests<br>GRGrok

Improve empty states<br>TP-163 Low<br>frontend<br>OCIn queue<br>Queue #2

1Tasks 2Agents 3Parallel 4Input 5Plans 6Safety 7Review 8Reports 9Online<br>CodexClaude CodeGrokOpenCodeOpenAI-compatible

SUPPORTED AGENTS<br>Bring the agents you already use.<br>cadre.rocks detects the coding CLIs installed on your machine and routes each task to the one you picked for that column. Same agents, same logins—now with a board, a queue and approval gates around them.

CC Claude Code<br>Anthropic

Anthropic's engineering agent in the terminal.<br>claudeCLI · BYOK

CO Codex<br>OpenAI

OpenAI's agentic coding CLI.<br>codexCLI · BYOK

GR Grok<br>xAI

xAI's assisted build CLI.<br>grokCLI · BYOK

OC Opencode<br>anomalyco

Open-source, multi-provider coding agent.<br>opencodeCLI · BYOK

{ } OpenAI Compatible<br>Local profile

Endpoint and model defined by a local OpenAI-compatible profile.<br>profileBYOK

Agents run locally through their own CLI or through your key. Adding a model never requires a new account with us.

MODELS & KEYS<br>Use the models you already pay for. No token markup.<br>Every agent, in every column, runs one of two ways: through the CLI login you already have, or through your own API key. Requests leave your machine straight for the provider—cadre.rocks never sits in the middle of them.

CLIUse the CLI login<br>No new billing<br>Point a column at a CLI that is already installed and authenticated. The run reuses that session.<br>✓Works with the subscription you already pay for<br>✓cadre.rocks detects which CLIs exist on this machine<br>✓Model and reasoning effort still chosen per column<br>✓No key to paste, nothing new to store

BYOKBring your own key<br>Your provider, your bill<br>Prefer API billing, a self-hosted endpoint or a model the CLI does not expose? Save a key for that agent.<br>✓Stored in the OS keyring—never in config files, logs or browser storage<br>✓Scoped to one agent in one column, separate from Chat BYOK<br>✓Any OpenAI-compatible endpoint through a local profile<br>✓Remove the credential and the column falls back to the CLI

Tokens are billed by your provider. cadre.rocks charges for the orchestration layer—boards, parallel runs, approvals—never per token.

Refresh models EXECUTION ORDER<br>1 Claude Code Anthropic<br>Local execution<br>CLI Uses the installed CLI loginBYOK Uses your saved API key<br>Model<br>CLI default<br>Reasoning<br>High

2 Codex OpenAI<br>Local execution<br>CLI Uses the installed CLI loginBYOK Uses your saved API key<br>Model<br>gpt-5-codex<br>Reasoning<br>Extra high

Execution credential configured in the system keyring.

AVAILABLE<br>GrokOpencodeOpenAI Compatible<br>Agents run in the order above. Without agents, the column stays manual.

Board settings → Column agents. The credential is scoped to that agent in that column, and is separate from Chat BYOK.

Choose your download Free local boards · Your CLI logins and keys stay on your machine

Agents execute.<br>You keep the gates.<br>Questions, plans, commands and code changes return to one decision surface before the workflow can continue.<br>Choose your download

How should board-wide approvals expire?<br>When the board closesAfter 24 hours Recommended

Implementation plan 1. Validate repository paths at the service boundary.<br>2. Refactor the import coordinator and add adversarial tests.

✓ command_policy.rs +64 −12 approval_service.rs +41 −18 command_policy.test.rs +23 −4

Start local. Go shared when the work does.<br>Begin with a private desktop board....