
A shared task board for people and their Claude agents, built as one claude.ai artifact. Views drawn
A shared task board for people and their Claude agents, built as one claude.ai artifact. Views drawn by code, not tokens.
·
English · 한국어 · 日本語 · 简体中文 · Español
One page is the board: tasks, threads, notices, a dependency graph and a code map. Keep it as a claude.ai artifact, or host it yourself for Claude Code and Codex.
The page draws everything from the board's database in code, so it stays live without spending a token, and agents read and write it through a few lean rules.
Install · See it · What you get · Use it · Self-host · How it works · For agents
Ask Claude. Paste this into Claude Code; it installs the skill and asks how many people, how you split work, the board language and where the first tasks come from:
Set up artiboard for me by following https://github.com/yulmucoffee/artiboard/blob/main/FOR_AGENTS.mdOr copy the skill yourself. Two commands, then say "set up a task board for this project" in a new session:
git clone --depth 1 https://github.com/yulmucoffee/artiboard.git "$HOME/.artiboard/src"mkdir -p "$HOME/.claude/skills/artiboard" && cp -R "$HOME/.artiboard/src/skills/artiboard/." "$HOME/.claude/skills/artiboard/"For Codex, copy to $HOME/.codex/skills/artiboard instead; for one repository, to .claude/skills/artiboard (or .agents/skills/artiboard) and commit. To upgrade, git -C "$HOME/.artiboard/src" pull --ff-only and copy again. Windows PowerShell commands are in FOR_AGENTS.md.
/plugin marketplace add yulmucoffee/artiboard
/plugin install artiboard@artiboardThe skill is then /artiboard:artiboard.
Open the live demo → A three-person team building a recipe app: the real board page on sample data, so claim a task, post in a thread, open the code map, switch to 한국어. Two agent updates arrive on their own.
Document view |
Kanban |
When a few people each run a few agent sessions on one codebase, the hard part is knowing who took what, what is blocked and what changed while you were away. artiboard is one live place for that, cheap for agents to use.
j/k keys, "new since you last looked" markers, light or dark.board_map instead of grepping, two open tasks on the same code warn each other, and the page draws it as nodes you open group by group.A board lives in one of two places. Both use the same page, documents and rules; the skill sets up either.
| claude.ai artifact | Self-hosted site | |
|---|---|---|
| Agents | Claude Code signed in with claude.ai | Claude Code, Codex, any MCP client |
| Data | The artifact's database | SQLite (one file) or Supabase |
| Runs on | claude.ai, nothing to run | This computer, your domain (Docker), or Vercel |
| Teammates join by | The artifact's Share menu | A token an admin issues |
| Boards | One artifact per board | One site, a board per project at /p/<project> |
Just ask, in any words and any language.
| You say | What happens |
|---|---|
| "Set up a task board for this project", "작업 보드 만들어줘" | setup: asks where the board lives, team size, split, language and task source; drafts tasks and the project's rules from your docs for a yes; publishes or creates the board, links your repositories and adds a short pointer to CLAUDE.md (and AGENTS.md). |
| "Add Cy to the board", "Add a QA status", "Switch the board to Japanese" | adapt: changes the board's config, or the page and the rules together. No republish. |
| "Move our board to our own server" | move: copies an artifact board's documents to a self-hosted project. |
| "Give me a board sign-in link", "보드 로그인 링크 줘" | A one-time browser sign-in link for a self-hosted board. |
| "Pick the next task", "Continue M1-4" | Daily work: board_start brings the board rules, the project's rules and your work; then claim, thread, summary, review, done. |
Share an artifact board from its Share menu with edit access; teammates' agents find it through the repository pointer. The first time a teammate picks their name under "I am", the board links it to their claude.ai account.
A site serves a project list at /, a board per project at /p/<project> and an MCP server at /mcp. It is the same page the artifact publishes, with a small runtime standing in for claude.ai. Needs Node 22.13 or newer.
On this computer (SQLite, data in ~/.artiboard/board.db):
cd "$HOME/.artiboard/src" && npm install --omit=devnode server/cli.js init --person ana --name Ana --project acme --title "Acme Board"node server/cli.js serveinit creates the first admin and prints a token with the commands that connect Claude Code and Codex. Open http://localhost:4100 and paste the token, or ask your agent for a sign-in link. For teammates on your network, serve with --host 0.0.0.0 and set BOARD_URL to the address they use.
On your domain with Docker (SQLite, TLS by Caddy): point a DNS record at the server, put your domain in docker-compose.yml, then:
docker compose up -ddocker compose exec board node server/cli.js init --person ana --name AnaOn Vercel with Supabase: create a Supabase project, then push the schema:
supabase link --project-ref <ref> && supabase db pushSet SUPABASE_URL, SUPABASE_PUBLISHABLE_KEY, SUPABASE_SECRET_KEY and BOARD_URL on the Vercel project, add your domain, vercel link, then npm run deploy:vercel (VERCEL_REGION=icn1 keeps the functions near a Seoul database). In Supabase Auth, add <BOARD_URL>/ to the redirect URLs. To upgrade a site, pull, supabase db push when there are new migrations, then deploy again.
Connect agents. Each person gets a token (an admin issues it in Members on the projects page, or with member add) and runs one of:
claude mcp add --scope user --transport http artiboard https://board.example.com/mcp --header "Authorization: Bearer abt_…"codex mcp add artiboard --url https://board.example.com/mcp --bearer-token-env-var ARTIBOARD_TOKENFor Codex, keep export ARTIBOARD_TOKEN=abt_… in your shell profile. Codex asks before every call of a tool that can overwrite documents; to let board_batch run without asking, add to ~/.codex/config.toml:
[mcp_servers.artiboard.tools.board_batch]
approval_mode = "approve"Code map. From a repository's checkout, build and upload its map; add --workflow to write a CI job for the default branch (with an ARTIBOARD_TOKEN repository secret), or --hook for git hooks in a checkout only you commit to, one per repository (Keeping the map current):
node "$HOME/.claude/skills/artiboard/scripts/codemap.mjs" --site https://board.example.com --project acme --repo acme/apiCustomize. Title, language and any UI string live in meta/config and change live. The page is per project: artiboard page set --project acme --file page.html for statuses, fields or views, page reset for the bundled page, which updates with the site.
| Setting | Default | Meaning |
|---|---|---|
BOARD_URL | http://localhost:4100 | The address people and agents use |
PORT, HOST | 4100, 127.0.0.1 | Where serve listens |
ARTIBOARD_DB | ~/.artiboard/board.db | The SQLite file |
SUPABASE_URL, SUPABASE_PUBLISHABLE_KEY, SUPABASE_SECRET_KEY | Keep the data in Supabase instead | |
ARTIBOARD_SITE | artiboard | The site's name on screens |
ARTIBOARD_LANG | en | Language of new boards and the site's screens (en, ko) |
ARTIBOARD_AGENT_PAGE_EDITS | off | 1 lets admins' agents change board pages through board_page |
artiboard serve [--port N] [--host H]
artiboard init --person ID [--name N] [--project ID --title T]
artiboard member add|token|link|revoke|access|join|remove|list
artiboard project add|repos|archive|restore|drop|list
artiboard page show|set|reset --project ID
artiboard import --project ID --file FILE [--map old=new]
artiboard map build|drop --project ID --repo RRun node server/cli.js with no command for every option. Admins do the same from the projects page (Members: add, token, sign-in link, access, revoke, remove), and everyone sets their own name and role in Profile. An admin puts a finished project away with Archive project in the ⋮ menu on its board (or project archive): it moves under Archived on the projects page, keeps everything and reads as before, but leaves the lists agents read and takes no writes until an admin restores it from the same menu. Delete project there (or project drop) removes the project and everything on its board for good. Access levels are admin, member (writes) and viewer (reads), and a site is one team: every member sees every project. Nobody gets a password: a token signs in an agent, and the page trades a token or a one-time link for a browser session.
| Tool | Does |
|---|---|
board_start {project or repo} | The board rules and the project's rules, new notices since your cursor, your open tasks, ready tasks, others' tasks waiting for review. With repo (a git remote) it finds the project |
board_claim {id, text, override?} | Claims a ready task, posts the claim, returns spec, acceptance, summary and the code the task names |
board_task {id} | Status, details, and the messages after the summary; brief leaves out the details |
board_post {task | tasks, text?, status?, branch?, pr?, start?, due?, repo?, summary?, touches?, override?} | A thread message plus changes to your task. done unblocks dependents, todo releases the task; without task, a notice; tasks moves several in one write. touches records the files you touch, so overlapping tasks warn each other |
board_map {explain, file, path, query, impact, inside} | The code map: explain a name, a file, a path between two, words, what depends on X, what X uses. No question: overview |
board_get, board_query, board_batch | Documents, as ArtifactData reads and writes them; tasks list key, title, status and owner unless fields asks for more |
board_page | The project's page: size, find, read (300 lines at a time, from line from), and for admins edits, html, reset |
board_projects, board_create_project, board_whoami, board_login_link | The site and its projects |
The everyday tools answer in short lines and check versions on the server, so a claim or an update is one call of under a hundred tokens.
flowchart LR
subgraph Artifact["claude.ai artifact"]
Page["Board page<br/>(views drawn by code)"]
DB[("Artifact database<br/>tasks · details · messages<br/>notices · cursors · meta")]
Page <-- live --> DB
end
People(["People"]) --> Page
AgentA(["Agent sessions"]) -- ArtifactData<br/>queries + batches --> DB
Repo["Your repo<br/>CLAUDE.md pointer<br/>docs/artiboard.md rules"] -. read on board work .-> AgentA| Rule | Effect |
|---|---|
| Code draws the board | Views, counts, readiness and "new" markers cost zero model tokens, at any team size. |
| Rules arrive with board work | Every session pays for a three-line pointer. The board rules and the project's short rules arrive only when board work starts. |
| Query, never list | Finding work is one query (status == todo, waitingOn == []); notices come as a delta after your cursor. |
| Small documents, short summaries | Task documents hold status; long text is opened for one task at a time; a handoff rewrites a five-line summary. |
| One batch per action | A claim is one call: task, message and cursor together, version-checked. |
| Self-repairing data | waitingOn is kept right by the page and by board_start, so agents never spend calls on bookkeeping. |
| Piece | What it holds | Who reads it |
|---|---|---|
tasks/<id> | Status only: owner, deps, waitingOn, claim, branch, PR, touches | The page; agents through queries |
details/<id> | Spec pointer, acceptance, a five-line summary | Agents, only for the task in hand |
tasks/<id>/messages | The thread: updates, questions, handoffs, blockers | Agents read the last eight; people read all |
notices | What everyone must know | Agents read only what is newer than their cursor |
cursors/<person> | When that person's agents last read notices | Agents |
meta/config | Title, people, language, repositories, any UI string | The page |
meta/rules, meta/notes | The project's own rules; shared status | People on the page; agents when board work starts, or when needed |
map/<repo> | A repository's code map: index and chunks | Agents through board_map; the page's Code map section |
docs/artiboard.md (repo, artifact boards) | The board rules, copied verbatim (a site gives them in board_start) | Agents, only in sessions that touch the board |
meta/config keys| Key | Example | Meaning |
|---|---|---|
title | "Acme Board" | Heading on the page |
subtitle | "What this board is for." | One line under the heading |
milestone | "M1" | Current milestone label |
lang | "en", "ko" | Built-in UI language. Others fall back to English. |
people | [{"id":"ana","name":"Ana","role":"backend"}] | Order sets colors. An optional uid pins a person to one claude.ai account. |
statuses | ["qa"] | Extra statuses, placed before done |
strings | {"v_graph":"Dependencies"} | Override any UI string, in any language |
footer | "..." | Replaces the footer line |
repos | [{"repo":"acme/api","label":"api"}] | The project's GitHub repositories. Agents find the board from their git remote; each task can name its own in repo. |
skills/artiboard/ the skill: SKILL.md, the board page (assets/board.html), the agents' references, the code map script
server/ the self-hosted site: HTTP app, MCP server, SQLite and Supabase stores, CLI, code map
api/, vercel.json, supabase/ a Supabase-backed site on Vercel: functions and migrations
Dockerfile, docker-compose.yml a SQLite site in a container, behind Caddy
examples/ the public demo, built from the page, demo-data.json and a map of examples/pantry
tests/ server tests (SQLite, and Supabase in CI), code map tests, the page in jsdom
FOR_AGENTS.md, AGENTS.md install steps for agents; how to work on this repositorynpm install && npm install --no-save [email protected] [email protected] && npm run checknpm run check runs every gate: tests on a throwaway SQLite file, the page in jsdom, and the copies that must stay in step. Issues and PRs are welcome; for a change to the skill, show how agent behavior or token use changes.
Built while coordinating a two-person, multi-agent product team on claude.ai artifacts. artiboard is independent and not affiliated with or endorsed by Anthropic. MIT License.
Paste this prompt into your agent. It reads this page and does the setup for you.
Read https://aiagentslisting.com/mcp/artiboard to learn what the "Artiboard" MCP server does and how to install it. Add it to my coding agent's MCP configuration as documented on that page, then confirm the server connects and list the tools it exposes.Agents can also browse this directory over MCP at https://aiagentslisting.com/api/mcp
claude mcp add --scope user --transport http artiboard https://board.example.com/mcp --header "Authorization: Bearer abt_…"This server doesn't publish a machine-readable tool manifest, and it doesn't expose a public endpoint we could ask. Connect the server locally to see what it exposes.
We could not run this server: its endpoint did not list its tools to an anonymous client, which usually means it asks for an API key or a sign-in. The four checks that need a live server are left out, and the grade tops out at B until we can read its tools.
Maintain this server? Claiming proves you own the listing and earns the last 4 points.
Claim this listingPaste this line near the top of your repository's README. The badge always renders the current score.
[](https://aiagentslisting.com/mcp/artiboard)Nothing comparable is listed yet.
No tagged releases on record.
Connect to the AI Agents Listing MCP server at https://aiagentslisting.com/api/mcp using the HTTP transport and complete the sign-in when it asks. Then call the submit_review tool for the listing "artiboard" with a rating from 1 to 5 and a short note on what worked and what did not.No reviews yet
Be the first to share your experience with this tool.
One email a week. New agents, MCP servers and skills, and what is actually getting traction.