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

setup-upapi

Skill0 starsChanged 9 days agoMITView on GitHub

Connects the upAPI MCP server to Claude and verifies it with one cheap call: where to connect in claude.ai, Claude Desktop, Cowork, or Claude Code, what the upAPI sign-in and consent screen asks, which tool table to expect, and how to recover when a tool is missing. Use when the user asks to 'set up upAPI', 'connect upAPI to Claude', 'install the upAPI connector', 'sign in to upAPI', 'upAPI tools are missing', or reports 'Tool not found' for an upAPI 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:

    Call upapi:text_analyze_post with { "text": "upAPI is connected." }.

    line 37
  2. Distinctive:

    Never ask the user for an upAPI password, API key, or access token in the chat.

    line 49
  3. Distinctive:

    Do not change upAPI settings or plans on the user's behalf.

    line 50

≈ “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/upapi/skills/setup-upapi/SKILL.md by scripts/sync-skills.mjs. Edit the plugin copy, then run the script. -->

Set up upAPI

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

  • MCP server key: upapi (tools are written upapi:<tool>)
  • MCP server URL: https://app.upapi.io/api/mcp?tools=claude (streamable HTTP, OAuth 2.1 with PKCE)
  • Docs: https://upapi.io/docs/mcp-clients

Where to connect

Claude Code with this plugin

  1. The plugin already registers the upapi server. Run /mcp, pick the upAPI server, and choose Authenticate, or just ask Claude to read a page and follow the sign-in link it opens.
  2. Without the plugin, add the server by hand: claude mcp add --transport http upapi 'https://app.upapi.io/api/mcp?tools=claude' (keep the quotes so the shell leaves the ? alone), 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

  1. upAPI is not in the Connectors Directory yet. Go to Settings > Connectors, select Add custom connector, name it upAPI, and paste https://app.upapi.io/api/mcp?tools=claude.
  2. Sign in to upAPI in the page that opens and approve the consent screen.
  3. On a Team or Enterprise plan, only an Owner can add a custom connector for the organization.

What the sign-in asks

upAPI asks you to sign in, then shows which app is asking and which workspace's plan and monthly quota its calls count against; with several workspaces you pick one. A guest session cannot connect (GUEST_FORBIDDEN); the user signs up for a full account first. Review and revoke connected assistants under Settings > Connected apps in the upAPI dashboard.

Verify the connection

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

Call upapi:text_analyze_post with { "text": "upAPI is connected." }. It runs without network access and costs 1 unit. If only upapi:search_ops and upapi:call_op are listed, call upapi:call_op with { "slug": "text-analyze.post", "input": { "text": "upAPI is connected." } } instead. A word count in the answer means upAPI is connected; say so and continue with the user's task.

When a tool is missing

  • Only upapi:search_ops and upapi:call_op (plus a few other tools) are listed: the connector was added with the plain URL https://app.upapi.io/api/mcp, which serves the compact table. Every skill works through upapi:call_op with the operation's slug. To get the named tools, re-add the connector with ?tools=claude.
  • Only upapi:search_ops is listed: the connection's grant does not include running operations. Disconnect and reconnect upAPI; if only upapi:search_ops is still listed, contact support@devino.ca.
  • ORGANIZATION_ACCESS_ENDED: the user left the workspace this connection was approved for. Reconnect and choose another workspace.
  • UNAUTHORIZED, an authentication error, or no upapi tools at all: the connector is not connected or the sign-in expired. Repeat the steps under "Where to connect".
  • AUTH_UNAVAILABLE: upAPI's sign-in check is briefly down. Wait and retry; do not reconnect.

Rules

  • Never ask the user for an upAPI password, API key, or access token in the chat. OAuth happens in the browser.
  • Do not change upAPI settings or plans on the user's behalf.
  • Setup makes only the one verification call above. Do not start a transcription or a capture 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.