AI assurance for finance teams
The review and sign-off layer between AI agents and your financial spreadsheets.
Explore AI assuranceBuilt on MCP (Model Context Protocol, the open standard AI tools use to connect to external systems), Rockhopper is the first spreadsheet governance platform you can drive directly from Claude, Microsoft Copilot, Endex, and Gemini - with full cell-level audit, true version-to-version comparisons, and AI agent transparency.
AI is the new finance dashboard.
Until now, if you wanted AI to help with your forecast or close, it had to piece together an answer from whatever you pasted into the chat. It was only as useful as the snapshot you gave it.
Rockhopper MCP changes that. Built on the open Model Context Protocol, it plugs your Rockhopper workspace directly into the AI tools your team already lives in - so the AI can audit, review, comment, and version your workbooks with full context.
Finance and accounting teams stop copy-pasting and start asking the questions that actually matter.
A full-time analyst, reviewer, and version controller - on call.
The AI only sees what your Rockhopper account is allowed to see. Every action is logged in your audit trail, just like every other action in the app.
Cell-level history surfaced in seconds, with the user, timestamp, and prior value.
"Who last changed C12 in the Q3 P&L Model, and what was it before?"
Uncommitted edits, grouped and explained - no more chasing analysts on Slack.
"Show me every unattributed change in our budget model this week and who made them."
Threaded, cell-anchored comments - written by your AI, owned by you.
"Add a comment on Sheet1!B5 asking why this assumption shifted 10%."
Kick off SOX-grade approval workflows without ever leaving the chat window.
"Create a review request for the latest version of the Q3 forecast."
Snapshot, branch, and roll back models with a single instruction.
"Create a new version called 'pre-board-edit'."
Clear the queue at the end of close - one conversation, every approval.
"List my pending review requests, then approve the one for the budget model."
One MCP endpoint. OAuth in your browser. Any client that speaks the Model Context Protocol can use it.
Connected in about two minutes.
Once your workspace is enabled for early access, paste the URL below into your AI client's MCP settings. Transport: Streamable HTTP.
https://mcp.rockhopper.co/mcp Your client opens a browser and signs you in just like any web app. No tokens to copy or store.
Try "List my Rockhopper files." Your enrolled workbooks come back inline.
Prefer a token?
Run the command below with a personal access token. Read-only or read & write - your call.
npx -y @rockhopper-co/mcp-server MCP traffic runs through the same audit, access, and compliance controls as everything else in your workspace.
Audited annually. The same controls that protect your Rockhopper workspace apply to MCP traffic.
The AI sees only what your account can see. A read-only token cannot post comments or open reviews.
Every MCP call lands in your Rockhopper audit log. Revoke tokens instantly from the same page.
Related resources
Keep exploring with practical guides, product pages, and proof points for your evaluation process.
The review and sign-off layer between AI agents and your financial spreadsheets.
Explore AI assuranceClaude for financial model intelligence, Rockhopper for enterprise governance.
See Claude + RockhopperEndex for model intelligence, Rockhopper for enterprise governance.
See Endex + RockhopperEarly access is rolling out to finance teams now. Tell us which AI client you use and we'll get you connected. Read the docs.
Request Access