AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Workflow Automation/
takt
takt logo

takt

Active·★ 1.4k·MIT·Updated 2026-10-27

TAKT (TAKT Agent Koordination Topology) is an open-source CLI that controls AI coding agents from the outside instead of trusting them to follow prompt rules. Planning, implementation, review, fix loops, human checkpoints, permissions and output contracts are defined in YAML workflow files, so the process is owned by the workflow rather than left to the agent.

Each step gets its own persona, policy, knowledge, instruction and output contract, which keeps context focused instead of letting one long session accumulate. Reviews cannot be silently skipped: findings route work back to fix steps, and human judgment can be requested. You describe a task in conversation, save it to a queue, run takt run to execute queued tasks in isolated worktrees, then use takt list to review diffs, merge, retry or requeue. On GitHub, takt caccia waits for CodeRabbit reviews and handles unresolved bot threads.

Providers include Claude Code, the Claude, Codex, OpenCode and Pi SDKs, the DeepSeek Harness SDK, Cursor, GitHub Copilot CLI and Kiro, so implementation and review can use different agents. Install with npm install -g takt and takt install for the SDK you want. README in English, Japanese and Chinese, with a written and video tutorial. MIT licensed, and built with TAKT itself.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-10-27.

License
MIT
Latest release
No GitHub releases
Last commit
2026-10-27
Contributors
35
Forks
116
Open issues and PRs
72
First published
2026-01-25
npm package
takt 0.73.0 · 4,277 downloads last week

Source: GitHub, npm and PyPI public APIs.

#agent-workflow#yaml#code-review#claude-code#codex#git-worktrees#cli
$ Install
$ npm install -g takt && takt install claude-sdk
↗ Visit site★ GitHub
01

Features

01YAML workflows with explicit plan, implement, review and fix steps
02Per-step persona, policy, knowledge, instructions and output contract
03Queued tasks executed in isolated worktrees with logs and reports
04Human checkpoints and per-step permissions
05CodeRabbit review loop with takt caccia
02

Why choose it

+Review and fix loops are explicit workflow transitions, so agents cannot skip them.
+Each step gets only the context it needs instead of one ever-growing session.
+Queued tasks run in isolated worktrees and leave logs and reports.
03

Trade-offs

!Writing and tuning YAML workflows takes upfront effort.
!SDK providers need a separate takt install step, and CLI providers need their own CLIs.
!Requires a Git repository with at least one commit.
04

Maintenance check

ActiveChecked: 2026-10-26
Last commit
2026-10-26
Commits, last 90 days
100+
Releases, last 90 days
0
Latest release
—
Open issues
63
Recent issues with at least one comment
1/14

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #1621 deepseek-harness: 未分類SDK/runtimeエラーを安全に表示する上流診断契約を整備する3 comments · 2026-09-27
  • #1673 takt director: 複数リポジトリの manager をまとめて扱う上位の層2 comments · 2026-10-03
  • #1905 system ステップで、workflow に固定したコマンドを実行できるようにしたい1 comments · 2026-10-26
  • #1776 本文を保護したまま、thinking中も独立した状態欄でスピナーを表示する1 comments · 2026-10-10
  • #1774 実行ログの write_stdin 表示が短い間隔で切り替わり、ちらつく1 comments · 2026-10-10

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

Node.js
CLI
Verified via docs
Claude Code / Codex / OpenCode
Providers
Verified via docs
Cursor / Copilot CLI / Kiro
Providers
Verified via docs
06

Quick start

1
$ npm install -g takt
2
$ takt install claude-sdk
07

Use cases

↳Enforce a review step that coding agents cannot skip
↳Queue several coding tasks and run them unattended
↳Mix different agents for implementation and review
08

Alternatives

⚡Top VS:
VSopenrig★8.3k→VScareer-ops★77.9k→VSAuto-claude-code-research-in-sleep★18.2k→VSloopx★6.4k→
openrig logo
openrig★ 8.3k
Build your own network of agents from Claude Code, Codex and Pi: persistent teams with roles, shared context and owned work.
VSCompare→
career-ops logo
career-ops★ 77.9k
Open-source AI job search agent and job finder: scan job boards, score each job 1-5 against your CV before you apply, tailor an ATS-friendly resume and cover letter, get interview prep and a job application tracker. It helps you fill in each application; you press Submit. Runs locally in your AI coding CLI (Claude Code, Codex, OpenCode and more).
VSCompare→
Auto-claude-code-research-in-sleep logo
Auto-claude-code-research-in-sleep★ 18.2k
ARIS ⚔️ (Auto-Research-In-Sleep) — Claude Code skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation via Codex MCP
VSCompare→
loopx logo
loopx★ 6.4k
Turn long-horizon goals into verifiable results. A local-first semantic control plane and research workbench for long-running agent teams, across sessions and runtimes.
VSCompare→
cindy logo
cindy★ 7.4k
Consider it done. The open-source AI agent that works out of the box · 想到,就能做到。开源、开箱即用的 AI Agent。
VSCompare→
harvey logo
harvey★ 2.8k
Autonomous AI sales agent powered by Claude Code or Codex. Finds prospects, writes cold emails, sends campaigns, handles replies, books meetings. No API keys needed — runs on your Claude or Codex subscriptions.
VSCompare→
agent-orchestrator logo
agent-orchestrator★ 13.8k
Open-source (Apache-2.0) orchestrator for Claude Code, Codex & 30+ coding agents. Plan, run and supervise agent teams from task to merged PR. Desktop (macOS, Windows, Linux) + iPhone & Android apps.
VSCompare→
kungfu logo
kungfu★ 4.5k
Your agents don’t hand off the work. Kungfu keeps the same Work moving across Codex, Claude, OpenCode, and your own execution surface.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
nrslib /
takt
↗

YAML workflows that make coding agents plan, implement, review and fix in steps they cannot skip

1.4k116MIT
Last commit1d ago
StatusActive
CategoryWorkflow Automation
Contributors35
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.