\n\n**Stop babysitting AI coding agents.**\n\nTAKT is an open-source CLI that turns AI coding agents into repeatable development workflows. Define planning, implementation, review, fix loops, human checkpoints, permissions, and output contracts in YAML, then run tasks with isolated worktrees and traceable logs.\n\nInstead of asking one agent to remember the whole process, TAKT gives each step its own role, context, and transition rules. Agents can code, but the workflow decides what happens next.\n\n\n\n- Run plan → implement → review → fix loops as explicit workflow steps\n- Keep context focused with step-specific personas, policies, knowledge, instructions, and output contracts\n- Execute queued tasks in isolated worktrees and inspect logs and reports afterward\n- Use Claude Code, Claude SDK, Codex SDK, OpenCode SDK, Pi SDK, the official DeepSeek Harness SDK, Cursor, GitHub Copilot CLI, or Kiro as providers\n\n**T**AKT **A**gent **K**oordination **T**opology orchestrates multiple AI agents with review loops, managed prompts, and per-step permissions.\n\nTalk to AI to define what you want, queue it as a task, and run it with `takt run`. Planning, implementation, review, and fix loops are defined in YAML workflow files, so the process is not left to the agent's discretion. TAKT coordinates Claude Code, Codex, OpenCode, Pi, the official DeepSeek Harness SDK, Cursor, GitHub Copilot CLI, and Kiro CLI as agents with different roles, permissions, and context.\n\nTAKT is built primarily for AI coding workflows, but the same model applies beyond coding: any task where multiple AI agents need to coordinate, or where review, judgment, and feedback loops can improve task quality.\n\nTAKT is built with TAKT itself (dogfooding).\n\n## Why TAKT\n\nAI coding agents do not automatically create a stable development process. In long-running work, they forget instructions, accumulate polluted context, blur implementation and review responsibilities, and often force humans to repeat the same feedback again and again.\n\nAdding more rules to prompts, `CLAUDE.md`, or skills can help, but it cannot enforce the process. Whether the rules are followed is still left to the agent's behavior.\n\nTAKT treats AI agents as something to be controlled from the outside, not simply trusted.\n\nWorkflows define the phases, and each step receives its own persona, policy, knowledge, instruction, and output contract. TAKT manages implementation, review, fix, and re-review flows declaratively. By separating responsibilities, knowledge, and constraints, then giving each agent only what it needs for the current step, TAKT improves task quality without bloating context.\n\nReviews cannot be silently skipped. Findings route work back to fix steps, and human judgment can be requested when needed. Tasks run in isolated worktrees, and each step leaves logs and reports so the path from task to PR remains traceable.\n\nTAKT runs all of this as a reusable agent process built from roles, phases, judgments, and feedback loops, so the development process stays reviewable and reproducible without constant human intervention.\n\n## Try It in 5 Minutes\n\nFrom a Git repository with at least one commit:\n\n```bash\nnpm install -g takt\n\n# Talk to AI, describe a task, use /go, then choose \"Queue as task\"\ntakt\n\n# Execute queued tasks in isolated worktrees\ntakt run\n\n# Review diffs, merge, retry, requeue, or delete task branches\ntakt list\n```\n\nIf this is your first run, configure a provider in `~/.takt/config.yaml` or use the API key environment variables listed in [Configuration](#configuration). SDK-based providers such as `claude-sdk`, `codex`, `opencode`, and `pi` can run with Node.js; `deepseek-harness` additionally requires the uv-managed environment created by `takt deepseek-harness install` on a supported platform; CLI-based providers require their external CLIs.\n\n### Video Tutorial\n\nFollow the [written tutorial](./docs/tutorial.md) with these hands-on walkthroughs:\n\n| Chapter 1 | Chapter 2 |\n|-----------|-----------|\n| [](https://youtu.be/HUcFFvOy39I) | [](https://youtu.be/UIlM2iM-rmA) |\n\n## TAKT vs Plain AI Coding Agents\n\n| Plain AI coding agents | TAKT |\n|------------------------|------|\n| The prompt asks the agent to follow a process | The YAML workflow owns the process |\n| Review steps can be forgotten or skipped | Review and fix loops are explicit transitions |\n| One long context keeps growing | Each step receives only the context it needs |\n| Implementation and review responsibilities blur | Personas, permissions, and output contracts separate responsibilities |\n| Work often lands directly in the current tree | Queued tasks run in isolated worktrees by default |\n| The path from task to result is hard to audit | Logs and reports preserve the path from task to PR |\n| The same process must be recreated by memory | Workflows are reusable, reviewable, and versionable |\n\n## Requirements\n\nTAKT requires Node.js `>=22.22.0`.\n\nThe provider you choose determines whether you need to install an external CLI or can run on Node.js alone via a TypeScript SDK.\n\nThese providers run via SDK (no CLI required, Node.js only):\n\n- `claude-sdk` — `@anthropic-ai/claude-agent-sdk`\n- `codex` — `@openai/codex-sdk`\n- `opencode` — `@opencode-ai/sdk`\n- `pi` — `@earendil-works/pi-coding-agent`\n\nThe `deepseek-harness` provider uses a managed environment that TAKT builds with `uv` and runs through a private JSON-RPC bridge. On a supported platform, run `takt deepseek-harness install` once before first use. npm install and npm lifecycle hooks never build this environment, and starting the provider during an install is unsupported because the provider does not wait for the installer lock.\n\nThe managed environment uses uv-managed CPython 3.12 and the fixed SDK/runtime versions declared in the shipped `pyproject.toml` and `uv.lock`. Linux x64/arm64 with glibc `>= 2.28` and macOS arm64 `>= 14.0` are supported; Windows, macOS x64, Linux musl, older Linux glibc, and older macOS fail fast, and TAKT does not silently fall back to another provider. A system Python installation is not required. Configure the standard `uv` network settings, such as `UV_INDEX_URL`, proxy, and certificate variables, when access to the package index requires them; TAKT passes those settings to the install command while `--locked` keeps the shipped lock authoritative. Install preflight requires `uv >= 0.11.0` and rejects a missing, unparseable, or older uv before deleting the existing managed environment.\n\nIf package-index access was previously configured with `pip`, migrate to uv's standard `UV_INDEX_URL`, proxy, and certificate environment variables; `uv sync --locked` uses the shipped lock as the dependency source.\n\nThe install `--python` option and provider `python_path` option have been removed because the managed environment is the only supported interpreter. Set `DEEPSEEK_API_KEY` and optionally `DEEPSEEK_BASE_URL` in the environment. This provider is a developer-preview compatibility surface; use the opt-in live smoke procedure in the configuration guide before relying on a new SDK/runtime pair.\n\nThese providers require an external CLI:\n\n- `claude` — [Claude Code](https://claude.ai/code)\n- `claude-terminal` — [Claude Code](https://claude.ai/code) driven in an interactive terminal session (also requires [`tmux`](https://github.com/tmux/tmux))\n- `copilot` — [GitHub Copilot CLI](https://docs.github.com/en/copilot/github-copilot-in-the-cli)\n- `cursor` — [Cursor Agent](https://docs.cursor.com/)\n- `kiro` — [Kiro CLI](https://kiro.dev/docs/cli/headless/)\n\nOptional:\n\n- [GitHub CLI](https://cli.github.com/) (`gh`) — for `takt #N` (GitHub Issue tasks)\n- [GitLab CLI](https://gitlab.com/gitlab-org/cli) (`glab`) — for GitLab Issue/MR integration (auto-detected from remote URL)\n- Java 17 or later (`java` on `PATH`) — for the model-checking stages of `/verify` in formal specification mode (`quint verify` and the Alloy Analyzer). The Quint CLI itself is bundled with TAKT; see [Formal Specification Verification](./docs/formal-verification.md) for what is downloaded on first use\n\n> **OAuth usage:** Whether OAuth is permitted varies by provider and use case. Check each provider's terms of service before using TAKT.\n\n## Quick Start\n\n### Install\n\n```bash\nnpm install -g takt\n```\n\nWith Nix flakes:\n\n```bash\nnix run github:nrslib/takt\nnix profile install github:nrslib/takt\n```\n\nThe Nix package installs the TAKT CLI itself. External CLI providers, `git`, and `gh`/`glab` still need to be installed and available on `PATH` or configured separately as described in [Requirements](#requirements).\n\n### Talk to AI and queue tasks\n\n```\n$ takt\n\nSelect workflow:\n ❯ 🎼 default (current)\n 📁 🚀 Quick Start/\n 📁 🛠️ Development/\n 📁 🔍 Review/\n\n> Add user authentication with JWT\n\n[AI clarifies requirements and organizes the task]\n\n> /go\n\nProposed task:\n ...\n\nWhat would you like to do?\n Execute now\n Create GitHub Issue\n ❯ Queue as task # ← normal flow\n Continue conversation\n```\n\nChoosing \"Queue as task\" saves the task to `.takt/tasks/`. Run `takt run` to execute — TAKT creates an isolated worktree, runs the workflow (plan → implement → review → fix loop), and offers to create a PR when done.\n\n```bash\n# Execute queued tasks\ntakt run\n\n# You can also queue from GitHub Issues\ntakt add #6\ntakt add #12\n\n# Execute all pending tasks\ntakt run\n```\n\n> **\"Execute now\"** runs the workflow directly in your current directory without worktree isolation. Useful for quick experiments, but note that changes go straight into your working tree.\n\n### Manage results\n\n```bash\n# List task branches — merge, retry, requeue, force-fail, or delete\ntakt list\n```\n\n## How It Works\n\nThe name TAKT comes from the German word for \"beat\" or \"baton stroke,\" used in conducting to keep an orchestra in time. TAKT uses **workflow** and **step** consistently in both user-facing and implementation-facing terminology.\n\nA workflow is defined by a sequence of steps. Use `steps`, `initial_step`, and `max_steps`. Each step specifies a persona (who), permissions (what's allowed), and rules (what happens next). Here's a minimal example:\n\n```yaml\nname: plan-implement-review\ninitial_step: plan\nmax_steps: 10\n\nsteps:\n - name: plan\n persona: planner\n edit: false\n rules:\n - condition: Planning complete\n next: implement\n\n - name: implement\n persona: coder\n edit: true\n required_permission_mode: edit\n rules:\n - condition: Implementation complete\n next: review\n\n - name: review\n persona: reviewer\n edit: false\n rules:\n - condition: Approved\n next: COMPLETE\n - condition: Needs fix\n next: implement # ← fix loop\n```\n\nRules determine the next step. `COMPLETE` ends the workflow successfully, `ABORT` ends with failure. See the [Workflow Guide](./docs/workflows.md) for the full schema, parallel steps, and rule condition types.\n\nReusable step definitions can be stored in `.takt/steps/` and expanded with `uses` before validation. See the Workflow Guide for fragment lookup and override rules.\n\nWorkflow files live in `workflows/` as the official directory name.\n\nWhen the same workflow name exists in multiple locations, TAKT resolves in this order: `.takt/workflows/` → `~/.takt/workflows/` → builtins.\n\n## Recommended Workflows\n\n| Workflow | Use Case |\n|-------|----------|\n| `default` | Standard development workflow. Scenario-based planning and test-first development with dynamic implementation companions, multi-perspective parallel peer review, adjudication, and a convergent fix loop. |\n| `maintenance` | A `default` variant for existing codebases: preserves contracts outside the change scope and limits work to causally related changes. |\n| `simple` | Lightweight workflow with the same minimal structure as `pure`; TAKT selects applicable domain facets per change, always including AI antipattern and architecture guidance. |\n| `pure` | Minimal workflow with no injected domain facets — trusts the model's own judgment and skill selection. |\n| `takt-default` | The workflow used to develop TAKT itself. Directly applicable to other CLI tool development. |\n| `takt-default-team` | A `takt-default` variant that runs implementation and remediation through Team Leader task decomposition. |\n| `review` | Multi-perspective review with dynamic reviewer selection and supervisor synthesis, without modifying code. |\n| `review-fix` | Multi-perspective review with dynamic reviewer selection, followed by the default workflow's adjudicated, verified remediation loop and final requirement check. |\n\nDomain-specific families (`simple-*` / `frontend` / `backend` / `dual` / CQRS / `*-mini` variants) remain available under the 📦 Legacy category.\n\nSee the [Builtin Catalog](./docs/builtin-catalog.md) for all workflows and personas.\n\n## Key Commands\n\n| Command | Description |\n|---------|-------------|\n| `takt` | Talk to AI, refine requirements, execute or queue tasks |\n| `takt exec` | Start instant Assistant/Worker/Review agent mode without writing workflow YAML |\n| `takt make` | Start Workflow Maker to create or revise an isolated workflow bundle interactively |\n| `takt add` | Refine a task through AI conversation and queue it (also from GitHub Issues) |\n| `takt run` | Execute all pending tasks |\n| `takt watch` | Monitor the task queue and auto-execute pending tasks (resident process) |\n| `takt ui` | Open the experimental Viewer-first Web UI on `http://127.0.0.1:20525` |\n| `takt list` | Manage task branches (merge, retry, requeue, force-fail, instruct, delete) |\n| `takt #N` | Use a GitHub Issue as the initial input for a task |\n| `takt eject` | Copy builtin workflows/facets for customization |\n| `takt workflow init` | Create a new workflow scaffold |\n| `takt workflow doctor` | Validate workflow definitions |\n| `takt workflow inspect` | Inspect a workflow's configuration and resolution sources |\n| `takt repertoire add` | Install a repertoire package from GitHub |\n\nSee the [CLI Reference](./docs/cli-reference.md) for all commands and options.\n\nThe Web UI defaults to Japanese. Use the language control in the header to switch\nbetween Japanese and English; the selection is saved in the browser for the next\nvisit. Viewer focuses on execution status, the observed execution path, live logs,\nand reports. Create a task opens the dedicated conversation surface for `/setup`\nand `/go`.\n\nTAKT also ships two client-integration entrypoints: `takt-acp` runs TAKT as an [Agent Client Protocol](./docs/cli-reference.md#acp-agent) agent over stdio JSON-RPC, and `takt-mcp` runs it as a stdio [MCP server](./docs/cli-reference.md#mcp-server) so an MCP client (Codex, Claude Code, …) can enqueue tasks, inspect task/run state, and send additional instructions to running worktree-clone tasks. Use `takt run` or `takt watch` to execute pending tasks.\n\nThe ordinary `takt` assistant conversation has the same read-only task-state view when its provider supports MCP. Use `/go` to turn a new task into an execution or queued task, and `/tell` to select and confirm an additional instruction for a running worktree clone.\n\n### Instant exec mode\n\n`takt exec` starts TAKT's interactive task-entry mode. The Assistant agent clarifies the request, `/go` turns the conversation into a generated workflow, Worker agent(s) implement the task, Review agent(s) review the result, the Replanning agent asks the user for direction when needed, and loop detection prevents repeated unproductive cycles.\n\nExec starts from the previous exec configuration, or the default configuration on first run; pass a preset name to start from that preset. Use `/setup` during the conversation to edit agents, loop detection thresholds, presets, and referenced facets. When `/go` runs, TAKT generates `.takt/exec/workflow.yaml` and executes it through the normal workflow engine; use `/cancel` to exit without running. Image attachments are supported while editing input (`/paste-image`, `Ctrl+V`, or an OSC 1337 inline-image paste).\n\nSee [Instant Exec Mode](./docs/cli-reference.md#instant-exec-mode) in the CLI Reference for preset resolution order, `/setup` persistence, image handling limits, and `session_key` behavior.\n\n## Configuration\n\nMinimal `~/.takt/config.yaml`:\n\n```yaml\nprovider: claude # claude, claude-sdk, claude-terminal, codex, opencode, deepseek-harness, cursor, copilot, kiro, pi, or mock\nmodel: sonnet # passed directly to provider\nlanguage: en # en or ja\n```\n\nRun metadata, sessions, traces, reports, and other run artifacts remain ordinary\nfiles under `.takt/runs//`. Resume and requeue preserve the applicable run\nstate and reports.\n\nBeyond these basics, `config.yaml` (legacy mode) supports internal-agent overrides (`takt_providers`) and `auto_routing`, which selects a provider/model per step from candidate pools with a `cost` / `balanced` / `performance` strategy. Auto-routing decisions can be recorded locally as NDJSON under `.takt/events/`; recording is opt-in (`takt telemetry enable` or `telemetry.routing_decisions`) and TAKT never uploads routing decisions. In runtime mode, provider/model/options and routing move to `runtime.yaml` (see below).\n\nOr use provider credentials directly (no CLI installation is required for claude-sdk, Codex, OpenCode, or Pi). DeepSeek Harness additionally requires the uv-managed environment created by `takt deepseek-harness install`:\n\n```bash\nexport TAKT_ANTHROPIC_API_KEY=sk-ant-... # Anthropic (Claude)\nexport TAKT_OPENAI_API_KEY=sk-... # OpenAI (Codex)\nexport TAKT_OPENCODE_API_KEY=... # OpenCode\nexport TAKT_CURSOR_API_KEY=... # Cursor Agent (optional if logged in)\nexport TAKT_COPILOT_GITHUB_TOKEN=ghp_... # GitHub Copilot CLI\nexport TAKT_KIRO_API_KEY=... # Kiro CLI\nexport DEEPSEEK_API_KEY=... # Official DeepSeek Harness SDK\n# Optional: export DEEPSEEK_BASE_URL=https://...\n# Pi uses its SDK credential store or provider-native environment variables.\n```\n\nSee the [Configuration Guide](./docs/configuration.md) for all options, provider profiles, and model resolution.\n\nOpenCode calls have a 60-minute provider-event inactivity limit by default: the\ntimer resets on each provider event, so a healthy call can run longer while\nevents continue to arrive. Raise the limit with\n`provider_options.opencode.guards.call_timeout_ms` (up to 86,400,000 ms). Guard\nprofiles and per-model overrides are described in\n[Provider inactivity deadline and OpenCode execution guards](./docs/configuration.md#provider-inactivity-deadline-and-opencode-execution-guards).\n\n### Dedicated provider configuration (`runtime.yaml`)\n\nProvider, model, provider options, auto routing, and internal-agent assignment\ncan live in a dedicated layer instead of `config.yaml`: `~/.takt/runtime.yaml`\nand `/.takt/runtime.yaml`, with the project layer winning. Workflow\nYAML has no provider/model/options/routing layer. Runtime mode is enabled by an\nactive `provider` section, not by the file existing; mixing an active\n`runtime.yaml` provider section with the legacy provider keys is rejected with\na diagnostic naming the file and the key to migrate to. Without an active\nprovider section, `config.yaml` behaves exactly as before.\n\nCompanion reviewers are disabled by default. Enable them with the top-level\n`companion.enabled: true` policy in `runtime.yaml`; global and project policies\ncombine with logical AND, so a global `false` cannot be re-enabled by a project\nsetting. See [docs/configuration.md](docs/configuration.md) for the schema, the\nmigration table, and companion policy details.\n\n## Customization\n\n### Custom workflows\n\n```bash\ntakt workflow init my-flow # Create a new workflow scaffold\ntakt workflow doctor my-flow # Validate a workflow definition\ntakt eject default # Copy builtin workflow to ~/.takt/workflows/ and edit\n```\n\n### Custom personas\n\nCreate a Markdown file in `~/.takt/facets/personas/`:\n\n```markdown\n# ~/.takt/facets/personas/my-reviewer.md\nYou are a code reviewer specialized in security.\n```\n\nReference it in your workflow: `persona: my-reviewer`\n\n`~/.takt/personas/` still works as a compatibility path, but `takt catalog` only scans the `facets/` directories.\n\nSee the [Workflow Guide](./docs/workflows.md) for details. The list of builtin personas is in the [Builtin Catalog](./docs/builtin-catalog.md).\n\n## CI/CD\n\nTAKT provides [takt-action](https://github.com/nrslib/takt-action) for GitHub Actions:\n\n```yaml\n- uses: nrslib/takt-action@main\n with:\n anthropic_api_key: ${{ secrets.TAKT_ANTHROPIC_API_KEY }}\n github_token: ${{ secrets.GITHUB_TOKEN }}\n```\n\nFor other CI systems, use pipeline mode:\n\n```bash\ntakt --pipeline --task \"Fix the bug\" --auto-pr\n```\n\nSee the [CI/CD Guide](./docs/ci-cd.md) for full setup instructions.\n\n## Project Structure\n\n```\n~/.takt/ # Global config\n├── config.yaml # Provider, model, language, etc.\n├── workflows/ # User workflow definitions\n├── facets/ # User facets (personas, policies, knowledge, etc.)\n└── repertoire/ # Installed repertoire packages\n\n.takt/ # Project-level\n├── config.yaml # Project config\n├── workflows/ # Project workflow overrides\n├── facets/ # Project facets\n├── tasks.yaml # Pending tasks\n├── tasks/ # Task specifications\n└── runs/ # Execution reports, logs, context\n```\n\nWorkflow definitions are stored under `workflows/`.\n\n## Adopting Spec-Driven Development\n\nTAKT enforces phase transitions declaratively as a YAML state machine, formalizes the artifact of each phase with output contracts, and routes deviations back via parallel review and fix loops. This structure suits Spec-Driven Development (SDD), where the spec stays at the center of the process. Once the spec is defined, the workflow enforces phase transitions, routes detected deviations back to fix steps, and does not complete until the verification gate passes.\n\nFor users who want to adopt SDD, the community provides [j5ik2o/takt-sdd](https://github.com/j5ik2o/takt-sdd) as a ready-made implementation. It ships workflows for Requirements → Gap Analysis → Design → Tasks → Implementation → Validation, plus an OpenSpec-style change-proposal flow. Install in one command:\n\n```bash\nnpx create-takt-sdd\n```\n\nSee [External Integrations](./docs/external-integrations.md) for other community integrations.\n\n## Documentation\n\n| Document | Description |\n|----------|-------------|\n| [Tutorial](./docs/tutorial.md) | Improve one example over three phases while queuing, running, and inspecting tasks |\n| [CLI Reference](./docs/cli-reference.md) | All commands and options |\n| [Configuration](./docs/configuration.md) | Global and project settings |\n| [Observability](./docs/observability.md) | Phase-level usage events and analysis workflow |\n| [Design Philosophy](./docs/design-philosophy.md) | Why TAKT is built around workflows, facets, feedback loops, and traceability |\n| [Workflow Guide](./docs/workflows.md) | Creating and customizing workflows |\n| [Builtin Catalog](./docs/builtin-catalog.md) | All builtin workflows and personas |\n| [Faceted Prompting](./docs/faceted-prompting.md) | Prompt design methodology |\n| [Token Saving](./docs/token-saving.md) | Measuring and reducing token consumption |\n| [Formal Specification Verification](./docs/formal-verification.md) | Checking agreed requirements with Quint and Alloy via `/verify` |\n| [Repertoire Packages](./docs/repertoire.md) | Installing and sharing packages |\n| [Task Management](./docs/task-management.md) | Task queuing, execution, isolation |\n| [CI/CD Integration](./docs/ci-cd.md) | GitHub Actions and pipeline mode |\n| [External Integrations](./docs/external-integrations.md) | Community examples that extend TAKT without modifying core (audit trails, etc.) |\n| [Changelog](./CHANGELOG.md) ([日本語](./docs/CHANGELOG.ja.md)) | Version history |\n\n### Simplified Chinese documentation\n\nSimplified Chinese documentation uses the `.zh-CN.md` suffix so it can coexist with the English and Japanese pages. Start with the [Chinese documentation index](./docs/README.zh-CN.md).\n\nTranslated coverage includes the onboarding path (README, tutorial, configuration, and CLI reference), workflow authoring, provider/external integrations, and task management. The remaining catalog, observability, design, prompting, token-saving, repertoire, CI/CD, testing, contributing, changelog, and internal design/development pages remain available in English or Japanese and are intentionally not duplicated here.\n\n## Sponsors\n\nTAKT is supported by [CodeRabbit](https://coderabbit.link/nrslib) through its Open Source Support Program.\n\n\n \n \n \n \n \n\n\n## Community\n\nJoin the [TAKT Discord](https://discord.gg/R2Xz3uYWxD) for questions, discussions, and updates.\n\n## Contributing\n\nPRs that accompany an issue are especially welcome. For bug fixes and small improvements, feel free to submit a PR without waiting for a maintainer response or for discussion to conclude.\n\nFor large refactoring or features that affect the project's design or compatibility, we recommend discussing the approach in an issue before implementation to avoid rework.\n\nSee [CONTRIBUTING.md](./CONTRIBUTING.md) for details.\n\n## License\n\nMIT — See [LICENSE](./LICENSE) for details.\n","readmeFilename":"README.md"}
![\"TAKT](\"./docs/assets/takt-logo-dark.svg\"){kind=logo}
\n ![\"npm](\"https://img.shields.io/npm/v/takt?label=npm\")
![\"GitHub](\"https://img.shields.io/github/stars/nrslib/takt?logo=github&label=stars\")
![\"CI\"](\"https://github.com/nrslib/takt/actions/workflows/ci.yml/badge.svg?branch=main\"){kind=icon}
\n ![\"license\"](\"https://img.shields.io/github/license/nrslib/takt\")
![\"Discord](\"https://img.shields.io/badge/dynamic/json?label=discord&query=approximate_member_count&url=https%3A%2F%2Fdiscord.com%2Fapi%2Fv10%2Finvites%2FR2Xz3uYWxD%3Fwith_counts%3Dtrue&suffix=%20members&logo=discord&logoColor=white&color=5865F2\"){kind=icon}
![\"TAKT](\"./docs/assets/tutorial-preview.gif\")
\n \n![\"CodeRabbit\"](\"https://victorious-bubble-f69a016683.media.strapiapp.com/Orange_Typemark_43bf516c9d.svg\")
\n