eyaltoledano/claude-task-master
An AI task-management loop that decomposes a spec into ordered, dependency-aware tasks and drives an agent through them one at a time across Cursor, Windsurf, and other runtimes. Agentiquette scores it 71/100 (A-tier), methodology v1.0, scored 2026-07-07.
Key facts
| Category | Workflow / loop tool |
|---|---|
| License | No license file at fetch time |
| Maintainer | eyaltoledano |
| Compatible agents | cursor, windsurf, claude-code, generic |
| File patterns | slash-commands |
| Triggering | slash-command |
| GitHub metrics | 27,771 stars · 2,612 forks · 201 open issues (fetched 2026-07-07) |
| Last commit | 2026-04-28 |
| Created | 2025-03-04 |
Editorial analysis
The best-known implementation of the plan phase of plan-execute-verify: it externalizes the plan as a task graph the agent cannot silently reorder. Task decomposition quality is genuinely good. The verify phase is thinner: task completion relies more on agent assertion than on evidence checks, which is the gap between this and a full closed-loop system. Maintenance cadence softened in the spring; worth watching.
Risks
License is a custom notice rather than a standard OSS license; review terms for commercial use. Ten-week commit gap at fetch time.
Evidence notes: Leverage: turns free-form agent work into bounded task execution. Reliability: strong plan discipline, weaker completion verification. Maintenance: 70-day commit gap at fetch, open-issue growth. Composability: works across major editor runtimes.
Install and first run
npm install -g task-master-aiAlternatives in this category
Spec-driven development workflow for Claude Code: requirements, design, tasks, and implementation as gated phases with approval between…
A minimal three-file task loop for AI dev agents: create a PRD, generate a task list, then execute one task at a time with human check-ins.
Compare with claude-code-spec-workflow · Compare with ai-dev-tasks
Embed this badge
Maintainers of scored entries can embed their live badge; it updates when the score does and links back to this page.
[](https://agentiquette.com/index/repos/claude-task-master)Frequently asked questions
Is eyaltoledano/claude-task-master good?
eyaltoledano/claude-task-master scores 71/100 (A-tier) on Agentiquette's methodology v1.0, scored 2026-07-07. Strengths and weaknesses are in the editorial analysis; the score is explained dimension by dimension in the scorecard.
What is eyaltoledano/claude-task-master for?
An AI task-management loop that decomposes a spec into ordered, dependency-aware tasks and drives an agent through them one at a time across Cursor, Windsurf, and other runtimes.
Is eyaltoledano/claude-task-master maintained?
Last commit 2026-04-28, 201 open issues at fetch time (2026-07-07). Maintenance is scored at 62 of 100.
Maintainer of eyaltoledano/claude-task-master? Claim this listing and suggest corrections.