mcpField
A
artifacts.boo

artifacts.boo

Remote
artifacts.booProductivity0Updated today
RepositoryVisit website

About artifacts.boo

artifacts.boo lets agents create small interactive pages for specific people, including reports, dashboards, polls, and checklists. Pages support per-viewer browser storage, live shared state, and anchored comments, with data isolated by team workspace.

Key features

  1. 01Create reports, dashboards, and polls
  2. 02Restrict access to specified viewers
  3. 03Support anchored comments and live shared state
  4. 04Provide per-viewer browser storage
  5. 05Isolate data by team workspace

Use cases

  1. 01Publish project updates for team review
  2. 02Collect votes or approval decisions
  3. 03Have members fill out interactive checklists
  4. 04Discuss specific content through anchored comments

MCP support

Agents connect to each team workspace's MCP server and act as a member. Tools support creating and editing artifacts; edits use patches rather than full rewrites.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://{workspace}.artifacts.boo/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

artifacts.boo

Agents publish small interactive pages (reports, dashboards, polls, checklists, mini apps) that specific people can open after verifying their email. Artifacts get per-viewer browser storage, live shared state and anchored comments. Each team gets a workspace at its own subdomain with its own Durable Object database, all on one Cloudflare Worker.

Try the hosted service

  1. Create a Free workspace with no card required. Free includes 20 artifacts and 3 members; guests are free. See pricing for all limits and the Team plan.
  2. Open your workspace's Settings for the connection steps for your MCP-capable agent. Each workspace has its own MCP address.
  3. Ask your agent to call artifact_guide, then create a private project update with a decision to review.

You do not need to deploy the app to use the hosted service. Help covers setup and feedback. New Free workspaces cannot turn on public links during their first day; start with a private artifact.

  • Agents connect to a workspace's MCP server at https://<slug>.<domain>/mcp, with OAuth or a personal agent token from the workspace's settings, and act as that member. Call artifact_guide first. Edits are patches (artifact_edit, artifact_patch_data), never full rewrites.
  • Viewers open /a/<id> in the workspace. They sign in once on the apex with an emailed code; the workspace then checks the artifact's share list.
  • Isolation: artifacts render in a sandboxed, opaque-origin frame with no network access. Storage and sync reach the frame only through the shell's postMessage bridge.

See docs/DESIGN.md for what it's for and the decisions behind it, and docs/DEPLOY.md to deploy and operate it.

Develop

bun install
bun run test        # unit tests
bun run typecheck
bun run build       # kit + worker
bun run deploy      # alchemy deploy (reads .env)
bun run e2e         # live end-to-end check in a throwaway workspace
bun scripts/seed-demo.ts   # publish demo/ to the public demo workspace
bun scripts/icons.ts       # re-render the PNG icons from their SVGs

The splash page's examples live in demo/: each folder is an artifact, and demo/examples.ts lists them with the votes and comments they start with.

License

FSL-1.1-ALv2: the source is available, and you may run, change and self-host it for your own team or organization, including for clients you do professional work for. You may not offer it to others as a competing hosted product. Each release becomes Apache 2.0 two years after it's published.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.