agentleFS
DevinoSolutionsDevinoSolutions/claude-plugins/skills/setup-demofy/SKILL.md

setup-demofy

Skill0 starsChanged 9 days agoMITView on GitHub

Connects the Demofy MCP server to Claude and verifies it with one read-only call: where to connect in claude.ai, Claude Desktop, Cowork, or Claude Code, what the Demofy sign-in and permission screens ask, and how to recover when a tool is missing or a permission was not granted. Use when the user asks to 'set up Demofy', 'connect Demofy to Claude', 'install the Demofy connector', 'sign in to Demofy', 'why can't Claude see my Demofy demos', 'Demofy tools are missing', or reports 'Tool not found' or SCOPE_MISSING for a Demofy tool.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    RATE_LIMITED or TOOL_FAILED: wait a moment and retry a read once.

    line 52
  2. Distinctive:

    Never ask the user for a Demofy password, API key, or access token in the chat.

    line 56
  3. Distinctive:

    Do not change Demofy settings, plans, or team members on the user's behalf while setting up.

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- Generated from plugins/demofy/skills/setup-demofy/SKILL.md by scripts/sync-skills.mjs. Edit the plugin copy, then run the script. -->

Set up Demofy

Connect the Demofy MCP server, confirm it answers, and fix the usual reasons a Demofy tool is missing. Sign-in and consent always happen in Demofy's own browser pages, never in the chat.

  • MCP server key: demofy (tools are written demofy:<tool>)
  • MCP server URL: https://app.demofy.io/api/mcp (streamable HTTP, OAuth 2.1 with PKCE and dynamic client registration)
  • Docs: https://demofy.io/docs/mcp

Where to connect

Claude Code with this plugin

  1. The plugin already registers the demofy server. Run /mcp, pick the Demofy server, and choose Authenticate, or just ask a Demofy question and follow the sign-in link Claude opens.
  2. Without the plugin, add the server by hand: claude mcp add --transport http demofy https://app.demofy.io/api/mcp, then authenticate from /mcp.
  3. A connector added on claude.ai is also available in Claude Code when you are signed in with the same claude.ai account, so there is no need to connect twice.

claude.ai, Claude Desktop, and Cowork

Demofy is not in the Connectors Directory yet. Add it as a custom connector:

  1. Open Settings > Connectors and select Add custom connector.
  2. Name it Demofy and enter https://app.demofy.io/api/mcp as the URL. Use this plain address; the ?tools=all form is for API-key clients, not for apps that sign you in.
  3. Select Connect, sign in to Demofy, and approve the permissions page.
  4. On a Team or Enterprise plan, an Owner may need to allow custom connectors first.

What the sign-in asks

Demofy asks you to sign in the way you usually do (Google, Apple, or email). It then shows the app asking and the permissions it wants, grouped: Read your data; Create and edit; Delete, publish or send; Billing and subscription; Profile, preferences, account deletion; Members and invites; Connected services; Support tickets; API keys. Every group the app asked for starts ticked. Untick anything you do not want; a tool whose permission you did not grant is never registered. Settings > Connected apps in Demofy lists every app you approved, with the workspaces it acts in, and disconnects it.

Making a demo needs "Read your data" and "Create and edit". Cancelling or deleting a demo needs "Delete, publish or send", and every destructive call still asks for a second, confirmed call.

Verify the connection

Make one read-only call and report what came back in a sentence.

Call demofy:demofy_list_demos if it is listed by name. If only demofy:search_tools and demofy:execute_typescript are listed (Code Mode, the default), call demofy:search_tools, then run demofy:execute_typescript with return await external_demofy_list_demos({});. It returns the newest demos in the workspace, or an empty list for a new account. Either answer means Demofy is connected; say so and continue with the user's task.

When a tool is missing

Most connections see Code Mode: tools/list shows only demofy:search_tools and demofy:execute_typescript, and every Demofy tool is called inside the sandbox as external_<tool>.

  • "Tool not found" on a named tool: switch to demofy:search_tools then demofy:execute_typescript instead of retrying the direct call.
  • ReferenceError: external_<tool> is not defined or SCOPE_MISSING: the permission behind that tool was not granted. Call external_request_scopes (when it is listed) and give the user the link it returns, or have them disconnect and reconnect Demofy and keep that group ticked.
  • WORKSPACE_REQUIRED or WORKSPACE_NOT_FOUND: the connection covers several workspaces. Call external_list_workspaces, ask which one, and include its workspaceId on every call.
  • An authentication error, or no demofy tools at all: the connector is not connected or was disconnected in Demofy (Settings > Connected apps). Repeat the steps under "Where to connect".
  • RATE_LIMITED or TOOL_FAILED: wait a moment and retry a read once; never retry a write without checking whether it landed.

Rules

  • Never ask the user for a Demofy password, API key, or access token in the chat. OAuth happens in the browser.
  • Do not change Demofy settings, plans, or team members on the user's behalf while setting up.
  • The verification call reads only. Do not create, edit, share, or delete a demo while setting up.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.