How Modeledge works

Modeledge is a clean web workspace and hosted MCP server for following companies, reading filings, searching documents, and organizing investment research. Connect Modeledge once to ChatGPT desktop, Claude Code, Codex, or another MCP-compatible AI client, and your agent can search the full text of SEC filings and earnings transcripts, compare disclosures, create and maintain valuation models, and save cited research back to your Modeledge account. You choose the AI client and model you prefer; Modeledge provides the company data, research context, and specialized tools.

Market intelligence through coverage that compounds. What you follow, model, and write accumulates, so next quarter starts from what you already know instead of from zero.

The system has four pillars. Each one builds on the last, and an agent connected over MCP can drive all of them.

Following a company is the core of market intelligence. Your followed companies are a coverage book, the names you want to stay current on.

• Your feed displays new filings and earnings transcripts for names you follow.
• Valuations can be set up to refresh after company and market updates.
• An agent can answer "what's new on my names?" in one call (GetFollowedCompanyActivity).

Follow companies you want to track and everything else builds up around this.

A finmodel is a structured, adjustable working view of a company's valuation. Some are quick first passes; others are researched and maintained over time. Assumptions are separated from arithmetic, and inputs, scenarios, and formulas derive the outputs. Excel in a language designed for agents. Every model has its own page (public or private).

Agents build and maintain finmodels over MCP (SaveFinModel, EvaluateFinModel), grounded in the filings and transcripts Modeledge indexes. You and other readers can turn the dials without touching the math and test a different view. Public models appear on the company's profile page under Models & Artifacts.

Binding a finmodel to a company you follow turns it into a tracked valuation: the model's headline outputs are stored on your valuation book, allowing you to sort by upside/downside and see the spectrum of prices versus your valuations.

A tracked valuation can be scheduled for review when material information arrives, making it a living view of what the company may be worth.

Agents manage tracked valuations over MCP (ManageTrackedValuation): list your coverage, check a valuation's status, bind or rebuild a model, and adjust assumptions.

Artifacts are durable investment notes and memos with their own web pages, saved from agent sessions or written directly. They compress the result of a long research session into context that you and your agents can retrieve later.

Agents save artifacts with CreateUserArtifact and search past work with SearchUserArtifacts, so earlier conclusions feed later ones.

Valuations age as companies change. Refresh valuations when the world changes — after earnings, on a filing event, or on a calendar you set — and each scheduled date carries an honest confidence level (guess, tentative, confirmed). A model's assumptions date shows what information it reflects; its updated date shows when the working document last changed. Every model also stores a refresh context: a runbook for how it's meant to be maintained, with methodology notes, known pitfalls, preferred sources, and a checklist.

Agents create and work these jobs over MCP (ManageValuationRefreshJob), so your models get revisited without you having to remember to.

The web app is free: create an account, follow companies, read filings and transcripts, and browse public models and research on company profile pages.

Every free account includes MCP access with a starter usage allowance. Connect your agent — ChatGPT desktop, Claude Code, Codex, Grok, or any MCP-compatible client — and it can drive all four pillars for you: follow names, read filings, build and maintain models, save artifacts, and schedule refreshes. Both sides share one account, so what the agent builds shows up in the app and what you follow in the app shapes what the agent sees. Paid Research plans provide higher usage limits for heavier workflows.

Once your agent is connected, start with a small read request to verify authentication, then move into your coverage workflow.

• Use Modeledge to find NVIDIA and return its ticker and CIK.
• What is new among the companies I follow?
• Summarize the latest earnings commentary for a company I cover.
• Save this research as a Modeledge artifact.