{"slug": "cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees", "title": "Cadre.rocks A local-first board for orchestrating coding agents in Git worktrees", "summary": "Cadre.rocks has launched a local-first board for orchestrating coding agents in Git worktrees, enabling parallel agent runs with human approval gates. The tool supports Claude Code by Anthropic, Codex by OpenAI, Grok by xAI, and Opencode by anomalyco, with no token markup and free local boards. It runs agents locally via CLI logins or BYOK, storing keys in the OS keyring, and offers optional online workspaces for team collaboration.", "body_md": "**3** agents running\n\n# Run coding agents like a team. Keep control of what ships.\n\nTurn 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.\n\nFree local boards · No credit card · Windows, macOS & Linux\n\n## The whole agent workflow, in one place.\n\nFollow a task from brief to branch, blockers, plan, diff and approval—then take the same work across devices when your team grows.\n\n01 / 09\n\n### Give agents a real brief.\n\nCreate tasks with descriptions, acceptance criteria, images and tags—then move them through the workflow your team already understands.\n\n- Custom sections and columns\n- Agent, tag and role routing\n- Context stays attached to the task\n\nShowing Payments: Give agents a real brief.\n\nSUPPORTED AGENTS\n\n## Bring the agents you already use.\n\ncadre.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.\n\n**CC**\n\n### Claude Code\n\nAnthropic\n\nAnthropic's engineering agent in the terminal.\n\n`claude`\n\nCLI · BYOK**CO**\n\n### Codex\n\nOpenAI\n\nOpenAI's agentic coding CLI.\n\n`codex`\n\nCLI · BYOK**GR**\n\n### Grok\n\nxAI\n\nxAI's assisted build CLI.\n\n`grok`\n\nCLI · BYOK**OC**\n\n### Opencode\n\nanomalyco\n\nOpen-source, multi-provider coding agent.\n\n`opencode`\n\nCLI · BYOK**{ }**\n\n### OpenAI Compatible\n\nLocal profile\n\nEndpoint and model defined by a local OpenAI-compatible profile.\n\n`profile`\n\nBYOKAgents run locally through their own CLI or through your key. Adding a model never requires a new account with us.\n\nMODELS & KEYS\n\n## Use the models you already pay for. No token markup.\n\nEvery 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.\n\n### Use the CLI login\n\nNo new billingPoint a column at a CLI that is already installed and authenticated. The run reuses that session.\n\n- Works with the subscription you already pay for\n- cadre.rocks detects which CLIs exist on this machine\n- Model and reasoning effort still chosen per column\n- No key to paste, nothing new to store\n\n### Bring your own key\n\nYour provider, your billPrefer API billing, a self-hosted endpoint or a model the CLI does not expose? Save a key for that agent.\n\n- Stored in the OS keyring—never in config files, logs or browser storage\n- Scoped to one agent in one column, separate from Chat BYOK\n- Any OpenAI-compatible endpoint through a local profile\n- Remove the credential and the column falls back to the CLI\n\nTokens are billed by your provider. cadre.rocks charges for the orchestration layer—boards, parallel runs, approvals—never per token.\n\n[Choose your download](#download-options)\n\nFree local boards · Your CLI logins and keys stay on your machine\n\n## Agents execute.\n\nYou keep the gates.\n\nQuestions, plans, commands and code changes return to one decision surface before the workflow can continue.\n\n**Needs input**\n\nHow should board-wide approvals expire?\n\n**Plan ready for review**\n\n**Implementation plan**\n\n1. Validate repository paths at the service boundary.\n\n2. Refactor the import coordinator and add adversarial tests.\n\n**Review changes**\n\n*✓*command_policy.rs\n\n**+64 −12** approval_service.rs\n\n**+41 −18** command_policy.test.rs\n\n**+23 −4**\n\n## Start local. Go shared when the work does.\n\nBegin with a private desktop board. Add online workspaces, cross-device sync, invites, and collaboration when the project becomes a team effort.\n\n### Local desktop\n\nBoards, repositories, worktrees and agent runs stay on your machine by default. Credentials remain in your system's secure storage.\n\n- ✓Local boards\n- ✓Local agent adapters\n- ✓Human approval gates\n- ✓Offline by default\n\n### Online workspaces\n\nKeep board activity in sync across computers and invite collaborators into the workspaces that need a shared view.\n\n- ✓Cross-device sync\n- ✓Workspace invites\n- ✓Shared board activity\n- ✓Team collaboration\n\nOnly the workspace data you choose is synced. Source repositories, worktrees and credentials remain local.\n\nGUIDES\n\n## Pick the part you care about. See it running.\n\nSix deep dives into the workflow, each with a live replica of the app doing the thing it describes—parallel runs, approval gates, command policy, local-first execution.\n\n6 GUIDES\n\n[ 1 CODING AGENT ORCHESTRATION](/coding-agent-orchestration/)\n\n### Agent orchestration\n\nThe full picture: boards, routing, parallel runs and approval gates in one operation.\n\n- 5\n- agent families in one queue\n- 16\n- concurrent runs on Team\n- 0%\n- markup on your tokens\n\n[ 2 MULTI-AGENT CODING WORKFLOW](/multi-agent-workflow/)\n\n### Multi-agent workflow\n\nRun several agents at once with isolated worktrees, a visible queue and ordered subtasks.\n\n- 1 → 16\n- concurrent runs by plan\n- 1:1\n- run to isolated worktree\n- 100%\n- of diffs reviewed by a human\n\n[ 3 CLAUDE CODE WORKFLOW](/claude-code/)\n\n### Claude Code\n\nKeep your Claude Code login. Add tasks, parallel runs, blockers and diff approval around it.\n\n- 0\n- new subscriptions required\n- 1 click\n- from question to resumed run\n- OS keyring\n- is where any key lives\n\n[ 4 CODEX CLI WORKFLOW](/codex-cli/)\n\n### Codex CLI\n\nStructured Codex runs with scoped command approvals, live logs and diff review.\n\n- 2 ways\n- to execute: CLI or BYOK\n- Per-run\n- branch and worktree\n- Backend\n- enforced command policy\n\n[ 5 HUMAN-IN-THE-LOOP](/human-in-the-loop/)\n\n### Human in the loop\n\nThe four gates: questions, plans, commands and diffs — each with its context attached.\n\n- 4\n- approval gates, not 400 prompts\n- Never\n- an automatic push or merge\n- Rust\n- backend enforces the policy\n\n[ 6 LOCAL-FIRST CODING AGENTS](/local-first-coding-agents/)\n\n### Local-first agents\n\nOffline by default, keys in the OS keyring, and sync you opt into workspace by workspace.\n\n- Offline\n- by default, no account needed\n- OS keyring\n- for every credential\n- Opt-in\n- sync, per workspace\n\n## Put your agents to work.\n\nKeep the final say.\n\nDownload cadre.rocks and move from scattered agent chats to tasks you can direct, inspect and approve.\n\nFree local boards · No credit card · Windows, macOS & Linux", "url": "https://wpnews.pro/news/cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees", "canonical_source": "https://cadre.rocks/", "published_at": "2026-08-13 18:58:10+00:00", "updated_at": "2026-08-13 19:13:46.838026+00:00", "lang": "en", "topics": ["developer-tools", "ai-agents", "ai-tools"], "entities": ["Cadre.rocks", "Anthropic", "OpenAI", "xAI", "anomalyco", "Claude Code", "Codex", "Grok"], "alternates": {"html": "https://wpnews.pro/news/cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees", "markdown": "https://wpnews.pro/news/cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees.md", "text": "https://wpnews.pro/news/cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees.txt", "jsonld": "https://wpnews.pro/news/cadre-rocks-a-local-first-board-for-orchestrating-coding-agents-in-git-worktrees.jsonld"}}