#9999th of 389 in Skills, plugins and prompts
comet
Resumable task workflows and Skill platform for coding agents
- Stars
- 3.3k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- JavaScript
0 on 12 May 20263,291 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Comet is an npm-installed CLI and Skill package that adds state-tracked, resumable workflows to coding agents. It offers a Native workflow and a Classic OpenSpec + Superpowers five-phase flow, plus a JavaScript/TypeScript runtime SDK, a Dashboard, memory and project knowledge, and an Eval command scoring Skills with rubric, Pass@k and Pass^k.
Who it is for: Developers running Claude Code or Codex on long, multi-phase tasks
Strengths
- Task state is kept in the project, so work resumes after interruption or session change
- Pure Node.js runtime runs on Windows, macOS and Linux without Bash or WSL
- Includes comet eval with rubric, Pass@k and Pass^k scoring, plus LangSmith/LangFuse integration
- Supervisor Changes split work into child changes run in isolated git worktrees
Weaknesses
- Many overlapping concepts (Native, Classic, Any, SDK, Eval, memory) make for a steep learning curve
- Classic mode depends on OpenSpec and Superpowers
- Benchmark figures come from the author's own 16-task experiment
- Requires Node.js 22.16+ or 24+
What it needs
- no GPU
- Needs Node.js 22.16+ or 24+, npm, Git
- Models: Codex, Claude Code
Quick answers
Is comet open source?
Yes. Its code is on GitHub under the MIT license.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |