惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

博客园 - 三生石上(FineUI控件)
S
SegmentFault 最新的问题
Jina AI
Jina AI
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
IT之家
IT之家
美团技术团队
V
Visual Studio Blog
人人都是产品经理
人人都是产品经理
博客园 - 叶小钗
有赞技术团队
有赞技术团队
GbyAI
GbyAI
宝玉的分享
宝玉的分享
腾讯CDC
M
MIT News - Artificial intelligence
博客园 - 【当耐特】
Google DeepMind News
Google DeepMind News
月光博客
月光博客
MyScale Blog
MyScale Blog
Last Week in AI
Last Week in AI
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 司徒正美
Recent Announcements
Recent Announcements
MongoDB | Blog
MongoDB | Blog

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
The Linear CLI Linear never shipped: driving your tracker...
Nico Acosta · 2026-06-27 · via DEV Community

If you searched for a Linear CLI, you already know the first surprise: there isn't an official one. Linear ships a hosted MCP server, a GraphQL API, and a TypeScript SDK, but no first-party command-line tool. What you find instead is a healthy little ecosystem of community CLIs, and a striking number of them were built specifically so an AI agent could drive Linear from the terminal.

That tells you something. The demand for a command line on top of an issue tracker is real, it is coming from engineers who live in the terminal, and increasingly it is coming from the agents those engineers run. This post is two things: an honest map of the Linear CLI options that exist today, and the argument for why the command line should be a first-class surface the tracker ships itself, not a wrapper you bolt on.

The Linear CLI options that actually exist

Linear's own developer surface is the GraphQL API and SDK. Everything below it on the command line is community-built. The well-known ones, as of mid-2026:

  • schpet/linear-cli: the most visible result, "Linear without leaving the command line." It ships a skill that helps AI agents use the CLI, which is a tell about who the audience is.
  • Linearis: a CLI "built for humans and LLM agents," with JSON output, smart ID resolution, and GraphQL queries tuned for agent use.
  • linctl: described by its author as "a Linear CLI purpose-built for agents like Claude Code."
  • Finesssee/linear-cli: a comprehensive CLI written in Rust, covering issues, projects, and cycles.
  • Composio's universal CLI: a connector layer that lets coding agents call Linear actions and triggers.

These are legitimate, and several are well made. If you use Linear and want a terminal workflow, one of them will probably get you most of the way there. The honest caveat is the one every community wrapper carries: it tracks an API it does not own. When Linear changes the GraphQL schema, the wrapper has to catch up, and you are depending on a maintainer's spare time for a surface that is load-bearing in your workflow.

Why "the CLI is community-built" is a structural fact, not an oversight

Linear is an excellent, GUI-first product. The keyboard craft is in the app: the command palette, the shortcuts, the speed of the interface. The command line was never the design center, so it landed in the community instead of in the box. There is nothing wrong with that choice. It is just a choice, and it has a consequence: the terminal is a second-class surface for Linear, supported by people who are not Linear.

Radial made the opposite choice. The CLI is a first-class surface, shipped and versioned with the product, sitting on the same REST API that everything else uses. It is not a wrapper around an API we reverse-engineered. It is the API, with a terminal on top.

What a first-class CLI looks like

The point of a command line on your tracker is that the work becomes scriptable. You file an issue without leaving the shell, you list your queue without a tab switch, and every command speaks --json so the same move pipes into anything:

radial create "Investigate 500s on checkout" -t ENG -p high --json

List what is on your plate the same way:

radial list --assignee me --status "in progress" --json

Because --json is on every command, the tracker stops being a place you visit and becomes a thing you script. CI can file an issue when a build breaks. A pre-commit hook can check what is assigned to you. And your agent can do all of it, because to the API there is no difference between you typing radial create and Claude Code calling the same endpoint over MCP. Every credential is a client of the API, not a billed seat, so the agents ride free.

That last part is the quiet through-line of the whole Linear-CLI ecosystem. Read the descriptions again: "built for LLM agents," "purpose-built for agents like Claude Code." People are not just asking for a terminal. They are asking for a surface their agents can drive. A tracker that ships the CLI and the MCP server itself, on one API, is meeting demand the community has been improvising against for two years.

Moving over is one command

If you are coming from Linear, the migration is a command, not a quarter. Export from Linear, preview the import dry first, then run it:

radial import --from linear export.json --dry-run

Drop --dry-run to bring your issues, projects, labels, comments, and history across for real. Your data is portable on the way out, too: one-command export is always available, so the CLI is not a trap, it is a door that swings both ways.

FAQ

Is there a CLI for Linear?

Not an official one. Linear ships a hosted MCP server, a GraphQL API, and a TypeScript SDK, but no first-party command-line tool. The community has filled the gap with several CLIs (schpet/linear-cli, Linearis, linctl, Finesssee/linear-cli, Composio), many of them built specifically for AI agents. They work, with the usual caveat that a community wrapper tracks an API it does not own.

Is a CLI outdated technology?

No, and the popularity of these agent-oriented Linear CLIs is the proof. A command line is the one interface a script and an agent can both drive without a human in the loop. As more of your workflow gets automated or handed to an agent, a scriptable, --json-speaking surface becomes more useful, not less. The GUI is for the human; the CLI is for everything else that needs to touch the tracker.

How is Radial's CLI different from a community Linear CLI?

It is first-party. The Radial CLI is shipped and versioned with the product, on the same REST API as the web app and the MCP server, so it does not lag behind a schema it does not control. A community Linear CLI is a wrapper a volunteer maintains against Linear's GraphQL API; when that API changes, the wrapper has to catch up.

Do my agents cost extra to use the CLI or API?

No. Every agent credential is a client of the API, CLI, or MCP server, not a billed seat, and there is no usage meter anywhere in the product. You pay one flat $50 per user, per year, billed annually, for the humans, and that price is locked at the rate you join. The Plain Software Pledge backs it: the day Radial ships a copilot, meters your usage, or charges you for AI you didn't ask for, your subscription is free.

The short version

Linear never shipped a CLI, so the community shipped a dozen, and most of them are pointed at agents. That is demand telling you where the terminal belongs: as a first-class surface the tracker ships itself, on one API, free for the agents to drive. That is the bet Radial made.

See the developer surface on the developers page, or read how Radial compares to Linear on the one thing a cap table won't let Linear match.


Originally published on the Radial blog.