By WellApp-ai
Grounded financial answers — expenses, runway, cash position, receivables, payables, FX exposure, reconciliation, and invoice drafting — computed live from Well's MCP financial graph.
Answer "who owes us money, and since when?" using Well's MCP financial graph — every outstanding customer invoice bucketed into standard aging bands (current, 1-30, 31-60, 61-90, 90+ days overdue), backed by real invoice data rather than guesswork. Use when the user asks "who owes us money", "accounts receivable aging", "AR aging", "outstanding receivables", "which customers haven't paid", "who's late paying us", or "overdue invoices owed to us". Requires a connected Well workspace with invoicing data and a resolvable `own_company`; if either is missing, this skill walks the user through connecting one or confirming their company first.
Show what bills are coming due and when, as a date-ordered cash-planning view of accounts payable with a running cumulative total, using Well's MCP financial graph. Use when the user asks "what bills are due", "upcoming payments", "what do we owe this week/month", "AP due dates", "what's our cash outflow looking like", "when are our bills due", or "payment calendar". Requires a connected Well workspace with invoicing/bills data; if none is connected, this skill walks the user through connecting one first.
Show how a company's cash balance has changed over time — a historical trend built from Well's MCP financial graph's real balance snapshots, never a forecast. Use when the user asks "cash balance trend", "chart our cash over time", "is our cash going up or down", "cash trajectory", "how has our cash changed", or "cash history". Requires a connected Well workspace with bank data covering more than one historical balance period; if none is connected or history is too thin, this skill walks the user through connecting one or says plainly there's not enough history yet.
Answer "how much cash do we have right now?" using Well's MCP financial graph — a fast, point-in-time snapshot of current bank/cash balances across all connected accounts, per currency, backed by real synced balances rather than guesswork. Use when the user asks "what's our cash position", "how much cash do we have right now", "current bank balance", "how much money is in the bank", or "what's our total cash on hand today". Requires a connected Well workspace with a banking connector; if none is connected, this skill walks the user through connecting one first.
Compose everything Well knows about one named company — customer or vendor — into a single 360 view using Well's MCP financial graph: profile info, contact channels, and the invoice relationship (as issuer, receiver, or both). Use when the user asks "build a customer 360 view", "customer 360", "give me a 360 view of X", "everything about [company name]", "vendor across every rail", "who is this company", "show me our history with X", or "tell me about our relationship with [vendor/customer]". Requires a connected Well workspace with at least one connector that has synced company/invoice data; if none is connected or the company can't be found, this skill says so instead of guessing.
External network access
Connects to servers outside your machine
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Grounded financial answers, delivered to you by your favorite agent.
You don't have to dig through ledgers, invoices, and bank feeds by hand. Ask, and get a grounded answer with the receipts attached.
Well is the financial operating layer for founders, finance leads, and lean teams. It connects to your bank accounts, accounting software, and invoicing tools, and gives your AI assistant secure, live access to that data through a standard called MCP — so it can answer with your real numbers instead of guessing.
No more exporting CSVs, copy-pasting numbers between tabs, or waiting on a bookkeeper to answer "how much runway do we actually have?" Well keeps your ledger, invoices, balances, and transactions synced and ready to query, in real time, from the tools you already use — Claude, Cursor, ChatGPT, and more.
https://api.wellapp.ai/v1/mcp
That's the address your AI assistant connects to. You'll add it once, during setup. (Jump to setup)
Connecting Well gives your AI assistant the tools to reach your data. On its own, that doesn't teach it how to use them well — what to check first, how to handle an account that's still syncing, or how to avoid blending three currencies into one meaningless number. That judgment is exactly what turns raw data into an answer you can trust.
This repo packages that judgment as Agent Skills — playbooks any AI assistant can follow. Each one confirms your account, checks there's enough real data to trust, pulls the right numbers, and — when something's missing — says so plainly instead of guessing.
What this saves you:
| Skill | Description | Details |
|---|---|---|
expense-breakdown | "Where does our money go?" — top expense categories and the biggest outstanding bills, from the ledger and invoices. | View details → |
runway-calculator | "What's our runway?" — real cash on hand vs. trailing burn, stated in months and days, with the full formula shown. | View details → |
cash-position | "How much cash do we have right now?" — a point-in-time snapshot across every connected account. | View details → |
cash-balance-trend | "Is our cash going up or down?" — a historical trend from real balance snapshots, never a forecast. | View details → |
accounts-receivable-aging | "Who owes us money, and since when?" — outstanding customer invoices bucketed by days overdue. | View details → |
bills-due | "What bills are due, and when?" — accounts payable sorted by due date with a running cash-out total. | View details → |
rank-clients-by-ltv | "Who are our best customers?" — customers ranked by total paid revenue to date. | View details → |
fx-exposure | "How exposed are we to currency risk?" — cash and invoices in foreign currencies, converted at real rates. | View details → |
company-profile | "Give me a 360 view of this company" — profile, contacts, and the full invoice relationship, customer or vendor. | View details → |
payment-invoice-lookup | "What happened with this payment?" — match a payment to its invoice, or list every unreconciled one. | View details → |
missing-receipts | "Which expenses are missing receipts?" — invoices with no source document attached, for compliance. | View details → |
draft-invoice | "Draft an invoice for this client" — creates a real invoice record in Well from a chat description. | View details → |
Connect Claude to your Well financial data — invoices, companies, contacts, transactions, and the accounting ledger — over a hosted OAuth MCP, with skills for real financial workflows.
npx claudepluginhub wellapp-ai/skills --plugin well-skillsPersistent coding companion for Claude Code, Pi, and Oh My Pi. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Enterprise AI agent orchestration plugin with 150+ commands, 74+ specialized agents, SPARC methodology, swarm coordination, GitHub integration, and neural training capabilities
Production-grade vibe coding suite. Structured workflows from discovery to deployment: start, build, ship, fix, refactor.
Memory compression system for Claude Code - persist context across sessions
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.