agentleFS
MosoFinMosoFin/mosofin-mcp-plugin/skills/connect/SKILL.md

connect

Skill3 starsChanged 22 days agoMITView on GitHub

Sign in to MosoFin and continue the user's request. Use when MosoFin tools are missing, a call returns Unknown tool or an authentication error, the MosoFin server needs authentication, the user asks how to connect or sign in to MosoFin, or the user says they finished signing in.

64 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use connect-apps, skill-share and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ArogyaReddySkill

    connect-apps

    Connect Claude to external apps like Gmail, Slack, GitHub. Use this skill when the user wants to send emails, create issues, post messages, or take actions in external services.

    Also in 19 projects that use this

  • ArogyaReddySkill

    skill-share

    A skill that creates new Claude skills and automatically shares them on Slack using Rube for seamless team collaboration and skill discovery.

    Also in 17 projects that use this

  • ComposioHQSkill

    agent-deep-links

    Build, validate, and troubleshoot deep links for Codex, Cursor, VS Code, Visual Studio, and similar tools. Use when users ask for clickable links (especially in Slack) that open threads, files, folders, or app settings.

    Also in 3 projects that use this

  • ComposioHQSkill

    helium-mcp

    Search real-time news with bias scoring, get live stock/ETF/crypto data with AI analysis, ML options pricing, balanced news synthesis, and meme search via the Helium MCP server.

    Also in 3 projects that use this

  • trendinuxSkill

    twitter-algorithm-optimizer

    Analyze and optimize tweets for maximum reach using Twitter's open-source algorithm insights. Rewrite and edit user tweets to improve engagement and visibility based on how the recommendation system ranks content.

    Also in 12 projects that use this

  • LangChainSkill

    langsmith-fetch

    Fetches LangSmith traces for debugging agent behavior. Use when troubleshooting agent issues, reviewing conversation history, or investigating tool calls.

    Also in 18 projects that use this

  • +1 more from ComposioHQ
  • +5 more from ArogyaReddy

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 6 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 55 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    Before any sign-in step, hold on to what the user asked: the question, workspace or company they named, period (resolve relative dates to …

    line 22
  2. Very specific:

    Search the session's tools for MosoFin's list_workspaces.

    line 29
  3. Very specific:

    If one exists, call list_workspaces.

    line 38
  4. Very specific:

    If the host offers its own MosoFin authentication tool (e.g.

    line 52
  5. Very specific:

    Create MosoFin — URL https://mcp.mosofin.com/mcp, Authentication

    line 83
  6. Distinctive:

    Start a new chat with the MosoFin app on, then ask again.

    line 85
  7. Distinctive:

    When the user says *done*, *signed in*, *continue*, or similar:

    line 96
  8. Distinctive:

    Search the tools again — MosoFin's data tools may have just appeared.

    line 98
  9. Very specific:

    Call list_workspaces once.

    line 99
  10. Distinctive:

    If the original request is no longer visible in this conversation, ask what they want to look at — never claim to remember a request you …

    line 107
  11. Distinctive:

    Ask for, accept, or repeat passwords, tokens, authorization codes, or callback URLs in the chat.

    line 120
  12. Distinctive:

    Treat a MosoFin sign-in as fixing a disconnected company file, or the reverse.

    line 128

“Common” is measured: kept by at least 6 in 10 of the 55 edited versions of this file we could read. “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

Sign in to MosoFin

MosoFin is a remote MCP server at https://mcp.mosofin.com/mcp using OAuth (DCR + PKCE). Sign-in is handled by the AI host you are in — the plugin ships no client id, secret, or local helper, and never handles tokens itself. This skill's job is to keep the user's request, give the one sign-in step for this host, and continue when they say done.

Signing in to MosoFin is not the same as reconnecting a company file. A disconnected QuickBooks (or other) file comes back as connected: false or datasource_not_active with a reconnect_url — that is handled by /mosofin:connections, not here. Say "sign in to MosoFin", never "refresh/reconnect the integration".

1. Keep the request

Before any sign-in step, hold on to what the user asked: the question, workspace or company they named, period (resolve relative dates to YYYY-MM-DD now), and the output they want. Sign-in must not cost them their question.

2. Check for working tools first

Search the session's tools for MosoFin's list_workspaces. Check the tool's origin rather than one exact prefix — any of these is MosoFin:

Where it came fromTypical id
This plugin in Claude Codemcp__plugin_mosofin_mosofin__list_workspaces
A MosoFin connector in Claude (claude.ai, Desktop, Cowork)mcp__MosoFin__list_workspaces or similar
ChatGPT / Codex appmosofin.list_workspaces

If one exists, call list_workspaces. If it succeeds, the user is signed in: go to step 4. Do not start a sign-in "to be sure". If two MosoFin sources both work, use one consistently; they may be signed in as different MosoFin users, so if their workspace lists differ, ask which account to use.

A loaded skill does not prove the connection works, and a missing tool does not prove the user's login expired. Report what you actually see.

3. Get the user signed in

Give only the next step for the host you are in — not every host's list.

Claude Code

  • If the host offers its own MosoFin authentication tool (e.g. authenticate on the mosofin server), call it once and show its URL as Sign in to MosoFin.
  • Otherwise: /mcp → plugin:mosofin:mosofin (shows Needs authentication) → Authenticate. The browser opens; return here after.
  • Browser can't open (SSH, container): from a shell, claude mcp login plugin:mosofin:mosofin --no-browser, open the printed link anywhere, and paste the final callback URL into that terminal — never into this chat.
  • plugin:mosofin:mosofin is not in /mcp: the plugin isn't installed or enabled here — /plugin marketplace add mosofin/mosofin-mcp-plugin, /plugin install mosofin@financehub, restart Claude Code.
  • Still failed right after signing in: some Claude Code versions hold a failed sign-in for about 15 minutes — Authenticate again in /mcp, or restart Claude Code. Never edit or delete auth or cache files.

Claude Desktop, Cowork, claude.ai

Settings → Connectors (or + in the prompt box → Connectors) → MosoFin → Connect. If MosoFin isn't listed, add a custom connector with URL https://mcp.mosofin.com/mcp, then Connect. Return to this conversation and make sure MosoFin is on for it. If sign-in lands on a localhost page the browser can't reach, the host's callback didn't complete — open the Connectors page and Connect again; don't retry the same link.

ChatGPT / Codex

  1. ChatGPT web → Settings → Apps & Connectors (or chatgpt.com/plugins).
  2. Turn on Developer mode if adding a custom connector (Settings → Security and login, or Apps → Advanced settings).
  3. Create MosoFin — URL https://mcp.mosofin.com/mcp, Authentication OAuth — and complete sign-in.
  4. Start a new chat with the MosoFin app on, then ask again. (ChatGPT can't add tools to an existing chat, so this is the one host where the user restates the request — offer a ready-to-paste version of it.)

Grok Build

Install mosofin from the xAI plugin marketplace and complete the sign-in it prompts for, then return here.

4. When the user comes back

When the user says done, signed in, continue, or similar:

  1. Search the tools again — MosoFin's data tools may have just appeared.
  2. Call list_workspaces once. Success → step 5; don't ask them to repeat their question.
  3. Still missing or unauthorized → say exactly what you see (e.g. "plugin:mosofin:mosofin still needs authentication") and give the one next step. Allow one retry in total; then stop and point them to support@mosofin.com with the visible, non-secret status. A "signed in" browser page alone does not prove access.

If the original request is no longer visible in this conversation, ask what they want to look at — never claim to remember a request you can't see.

5. Continue the request

Workspace confirmation is still mandatory after sign-in: read the workspace name(s) back and get an explicit yes (see /mosofin:workspaces). Then carry on with the held request through /mosofin:query-workspace or the matching skill. If the user only asked to connect, say they're signed in, name the workspaces they can see, and suggest one useful first question.

Never

  • Ask for, accept, or repeat passwords, tokens, authorization codes, or callback URLs in the chat. If a user pastes one, tell them not to and don't use it.
  • Build OAuth or login URLs yourself, call MosoFin's HTTP endpoints directly, or write MCP configuration to work around a missing connection.
  • Loop through repeated sign-ins, reinstalls, or new chats.
  • Invent workspaces, companies, or figures while signed out, or answer from memory as if it were live data.
  • Treat a MosoFin sign-in as fixing a disconnected company file, or the reverse.

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.