Introduction
The home for AI-generated work: agents publish interactive artifacts, humans review them with a link, and good feedback becomes the next version.
Artifact Cafe is the home for AI-generated work. Every agent produces artifacts: dashboards, plans, specs, reports, prototypes. Today they scatter across chats and markdown files, then disappear as fast as agents make them. Artifact Cafe gives them a shareable home: agents publish an artifact from the CLI, humans review it through a link with no account, and good feedback becomes the next immutable version.
When you want to iterate with an agent while actively reviewing, you can opt into a live editing session:
npx artifact-cafe@latest publish . # → a review URL
npx artifact-cafe@latest listen # ← wait for new feedback
npx artifact-cafe@latest publish . # → the next versionGet started: hand it to your agent
You don't have to run anything yourself. Paste this into your agent (Claude Code, Codex, Cursor, or any tool that can run a shell command) and let it set up and publish:
Install the artifact-cafe skill:
npx skills add artifact-cafe/skill --skill artifact-cafe -g
(without npm: curl -fsSL https://artifact.cafe/install.sh | bash)
Then publish this project with `artifact-cafe publish .` and send me the review link — reviewers can comment on it without an account.
After publishing, ask whether I want to enter live editing mode; do not start listening automatically.Prefer to run it yourself? Follow the Quickstart for the manual path.
Manifesto
AI makes drafts. Humans brew ideas.
A bean by itself isn't coffee; context by itself isn't an answer. Great artifacts aren't generated: they're extracted. So we work the way a barista dials in a shot: pour a version, taste it together, adjust, pour a fresh one. The cup is never edited, and the notes stay with the cup they described.
The vision
Generating work is no longer scarce. Every artifact still needs review, discussion, revision, and a place to be remembered. Artifact Cafe bridges AI generation and human decision-making: a durable, interactive home where work lives, gets better, and gets reused, from one cup to a whole cafe.
Keep reading
Quickstart
The publish → review → v2 loop in three commands.
For your agent
Install the skill once, no CLAUDE.md paste needed.
Publishing concepts
Entry files, limits, immutable versions, and slugs.
Reviewing
How guests comment: point and text feedback, no login.
Workspaces
Team home for artifacts: members, roles, folders, and visibility.
Live agent loop
Let an agent listen for comments, revise, and publish again while review stays open.
Markdown for agents
Fetch any artifact as Markdown, by suffix or content negotiation.
Password protection
Gate a review link behind a shared password.
CLI reference
Every command, flag, and --json output.
HTTP API
The public /api/v1 the CLI and agents call.
Error reference
Stable codes returned by the API.
Security and trust
How untrusted artifact HTML is isolated.
View stats
Owner-only views, unique visitors, and time on page.
Prefer plain text? The same operating notes live at /docs.md, and an agent-oriented summary at /llms.txt.