Workspace beta is live — BYO-LLM chat wired to 57 SEC tools. Try it free →
ValueinValuein
MCP Server live · Workspace beta open

Point-in-time data. Automated workflows. AI agents doing the heavy lifting.

Look-ahead-bias-free SEC fundamentals your team and your AI agents turn into alpha — find the edge, dodge the blowup, prove the call. Institutional-grade, at under 5% of a terminal.

More alphaFewer blowupsA fraction of the cost

Free S&P 500 tier — no credit card · one token across MCP, SDK & Workspace

Spread, monitor, and defend — in minutes, not days.

Standardized, point-in-time financials that drop straight into your model — every number traced to the filing it came from.

Works with your existing stack

Claude
Cursor
Codex
Windsurf
ChatGPT
Python
DuckDB
Parquet
Seconds
To spread any filing
Zero
Look-ahead bias
19,000+
Companies, incl. delisted
<5%
Of a Bloomberg seat

What we do, simply.

We turn raw SEC EDGAR filings into clean, point-in-time financial data — and deliver it wherever you work.

01

We ingest every SEC filing

10-K, 10-Q, 8-K, 20-F and more — every U.S. public company since 1993, including the ones that went bankrupt.

02

We standardize and timestamp it

Thousands of messy XBRL tags become clean, comparable concepts — each stamped with the moment it went public, so there's no look-ahead bias.

03

You use it anywhere

Through your AI agent (MCP), your browser (Workspace), your code (Python SDK), or bulk Parquet — all on one token.

Channel 1 · For your AI agent

Put a tireless analyst on every name you cover.

In 2025, developers queried APIs. In 2026, AI agents query MCP servers. Add mcp.valuein.biz/mcp to Claude, Cursor, Codex, Windsurf, ChatGPT — anything speaking the protocol — and every tool call returns SEC-filed data, point-in-time accurate.

mcp.valuein.biz/mcp
R
What's NVDA's FY2024 revenue and operating margin?
Claude Desktop·Sonnet 4.5

Calling Valuein MCP…

search_companiesNVDA → CIK 1045810
get_company_fundamentalsFY2024 · 10-K
get_financial_ratiosmargins · cash conversion

One-line connection

Add mcp.valuein.biz/mcp to any MCP-compatible client. Claude Desktop, Cursor, Codex CLI, Windsurf, custom agents — that's the entire integration.

57 tools · 22 SOPs

Fundamentals, ratios, peer comps, smart-money flow, screening, forensic audit, DCF compute, document generation — across 12 categories.

Token-scoped, edge-enforced

Bearer auth with tiered permissions. Free S&P500 tier. Pro for the 19,000-entity universe. Institutional adds redistribution + webhooks.

Edge-computed queries

Distributed Cloudflare edge — queries run milliseconds from the data, no server round-trips. <80ms median, <450ms P99.

Try it live — real MCP calls, no signup

Try it live

● Sample tier · no token

Real SEC data, one click away.

These are real MCP tool calls running against mcp.valuein.biz — no signup, no key.

Want the full universe?

Open the MCP playground
Response · get_company_fundamentals

Pick a query on the left to call the MCP server. Real response will appear here.

Plug it into Claude Desktop · Cursor · Codex CLI · Windsurf

Claude Desktop
Connected to Valuein MCP · 57 tools
1 · Connect
claude_desktop_config.json
2 · Ask
Best prompts & SOPs
3 · Answer
Structured, sourced

Drop the Valuein block into your config — same pattern in Cursor, Codex CLI, Windsurf, and any MCP-compatible client.

~/Library/Application Support/Claude/claude_desktop_config.json✓ saved
{
  "mcpServers": {
    "valuein": {
      "url": "https://mcp.valuein.biz/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_VALUEIN_TOKEN"
      }
    }
  }
}
Restart Claude Desktop57 tools · 22 SOPs · 3 resourcesSame token unlocks Workspace + SDK
Channel 2 · For your browser
BETA

Go from filing to defensible thesis in an afternoon.

Same MCP, our chrome. Bring your own Anthropic or OpenAI key — sealed in a 24h cookie, zero retention — and ship theses, alerts, and shareable reports from one tab. Free tier opens it up to the S&P500 universe; Pro widens to the full 19,000+.

valuein.biz/workspace
Agents
AI Chat
Workflows
Tools & SOPs57 + 22
Research
Feed
Theses
Watchlists
Reports
Alerts
R
Equity research brief on NVDA, forensic depth.

Running flagship equity_research_brief SOP…

search_companiesresolve NVDA → CIK 1045810
get_company_fundamentalsrevenue · margins · cash flow
get_financial_ratiosROIC, FCF margin, leverage
forensic_auditaccruals · revenue quality flags
generate_research_brief_docxrendered → /r/nvda-2026-q1
Ask anything — wired to 57 SEC tools…

BYO-LLM Chat

Bring your own Anthropic or OpenAI key, sealed in a 24h httpOnly cookie. Zero retention. Wired to all 57 MCP tools.

22 agentic SOPs

One-click flagship workflows — equity research brief, capital allocation, smart-money screen, forensic audit, peer comps.

Claims → Theses

Atomic, evidence-backed claims — cited to a filing, auto-graded against the data — composed into theses with conviction + horizon. Plus 500-ticker watchlists.

Reports + Profile

Tiptap editor with autosave + version history. A verifiable claim + thesis track record, private by default — with secure sharing on your terms.

Plus:WatchlistsEmail alertsWebhook alertsDaily briefBYO-LLM
Channel 3 · For your backtest

Backtests that can't see the future.

Crunch millions of rows out-of-core — pip install valuein-sdk and let DuckDB stream R2 Parquet locally. 54 prebuilt SQL templates, point-in-time enforcement baked into every row, no API round-trips. Your Sharpe survives out-of-sample.

fundamentals.py
# pip install valuein-sdk
# Token required for every tier except Sample.
# echo 'VALUEIN_API_KEY="your_token"' >> .env

from valuein_sdk import ValueinClient, ValueinError

try:
    with ValueinClient() as client:
        try:
            # Build & run a raw SQL query
            sql = "SELECT COUNT(cik) FROM entity"
            result = client.run_query(sql)
            print(result)

            # Or use a prebuilt SQL template with params
            df = client.run_template(
                "fundamentals_by_ticker",
                ticker="AAPL",
            )
            print(df)
        except ValueinError as ve:
            print(f"Query failed: {ve}")
except Exception as e:
    print(f"Initialization failed: {e}")
  • Native DuckDB + Parquet: 111M+ facts query locally in milliseconds
  • Point-in-time accepted_at timestamps on every row — no look-ahead bias
  • Full universe — 19,000+ tickers including all delisted entities
  • From pip install to DataFrame in under 60 seconds

SEC fundamentals. As reported. As they happened.

We don't restate. We don't adjust. Every number is exactly what the company filed with the SEC, timestamped to the moment it became public knowledge. Your backtest sees what the market saw — nothing more, nothing less.

PIT

Point-in-Time. Not point-in-hindsight.

Every fact is recorded with its EDGAR acceptance timestamp. Query any date and get exactly what was public knowledge at that moment. No look-ahead bias. No restated surprises. Your 2022 backtest will never accidentally see 2024 amendments.

FULL UNIVERSE

The dead companies are still here.

Enron. Lehman Brothers. Every delisted, merged, and bankrupt entity that ever filed with the SEC is in our dataset. Your model trains on the full universe — not just the winners that survived.

8-K INCLUDED

Don't wait for the quarter.

8-K current reports capture material events as they happen — CEO departures, acquisitions, unexpected write-downs. Most datasets skip these. We parse every one, giving event-driven models a latency advantage over quarterly-only feeds.

1993–TODAY

Dot-com. Financial crisis. COVID. All of it.

111M+ facts dating back to 1993. Deep enough to train ML models across multiple full market cycles and stress-test strategies against every regime shift in modern financial history.

Built for

AI Agents & Copilots

Catch a restatement or covenant breach the moment it files — your agent calls verified SEC tools and never hallucinates a number, because every figure traces to the filing.

Financial Analysts

Spread any 10-K in seconds, monitor your whole watchlist, and ship an IC-ready thesis — every number traceable to its filing. No terminal required.

Quantitative Researchers

Backtest across 30 years and every delisted name, point-in-time — so your Sharpe survives out-of-sample instead of overfitting to the survivors.

Portfolio Managers

See the smart money and the red flags first — 13F flows, insider clusters, and forensic scores across the whole book, with alerts the moment something moves.

Token-scoped access. Edge-enforced.

Every request is authenticated at the edge. Your token determines data scope and rate limits — checked on every call, the same across MCP, Workspace, and SDK.

TierData ScopeRate LimitPrice
S&P500S&P500 · 500+ tickers · 1993–present60 req/min · 1,000 req/hrFree
ProActive + delisted US universe · 19,000+ entities · 15-year history (2011→present)100 req/min · 3,000 req/hr$49/mo
InstitutionalTwo datasets: fundamentals (111M+ facts, 19,000+ entities, 1993–present) + smart-money (~78M rows · 6 tables: Forms 3/4/5/144 + 13F/13D/13G)300 req/min · 10,000 req/hr$499/mo
Get Your Token

S&P500 tier — free, no credit card required.

One token. Three ways in.

Connect Claude or Cursor in 30 seconds — or try the new Workspace in your browser. Free S&P500 tier includes full 1993–present history. No credit card required.

Free S&P500 tier — Pro from $49/moNo credit card requiredMCP · Workspace · Python SDK — one token