For AI assistants

DEOS Open and CULT+MATH research, inside your assistant.

Start with DEOS Open. On the full server, list_standards lists the free DEOS Open standards and get_standard returns one in full with its template. An agent picks the standard that fits the decision, reads it and follows its steps. When a step needs evidence, the research and paid runs below supply it. What DEOS Open is.

CULT+MATH publishes sourced, dated research on executive decisions: studies and short answers on cybersecurity, games, streaming, film, consumer packaged goods, ERP software, private equity and AI visibility. The same research is available to Claude, ChatGPT and any client that takes a remote MCP server, so an assistant can find a finding, read the page behind it, and cite it with its date and limits.

No account, no key, no sign-in. Paste an address below into your assistant and ask your question.

Three addresses, one server

  • cultmath.com/mcp/chatgpt: DEOS Open and the free research. Read any DEOS Open standard, search the studies and answers, read one in full.
  • cultmath.com/mcp/claude: DEOS Open and the free research, plus what each Decision Brief covers and its page.
  • cultmath.com/mcp: everything above, plus buying a single decision and paid measurement runs that an assistant's own wallet pays for.

Updated November 7, 2026
Model Context Protocol · Streamable HTTP

Connect it

ChatGPT

Install CULT+MATH Research from ChatGPT's app directory. It is the same free research described on this page, read-only, with no authentication.

Install it in ChatGPT

To add it by address instead, turn on developer mode in ChatGPT's settings and add an app with the address below. The listing and the address reach the same server.

https://cultmath.com/mcp/chatgpt

Claude

In Claude, open Customize, then Connectors, and add a custom connector with this address. Custom connectors are available on Claude's paid plans.

https://cultmath.com/mcp/claude

Any other MCP client

Cursor, Claude Code, Windsurf and other clients take a remote server by address. Use the full server if you want the paid runs as well.

https://cultmath.com/mcp

The server answers JSON-RPC over HTTP POST, statelessly (no session id). Its server card is at /mcp/server-card.

LangChain or the Vercel AI SDK

Install the tools as a package. The free research tools work with no setup. Paid runs return their price first and only charge through an x402 payment fetch you pass in, capped at $1.00 a call unless you raise it. The package never sees your wallet key.

npm install @cultmath/agent-tools

Package page and examples: npmjs.com/package/@cultmath/agent-tools.

The tools

Every tool in the Claude and ChatGPT editions is read-only: it looks things up and changes nothing. On the full server, buy_decision_evidence is the one exception, because it starts a checkout. Each result names its source page and the credit to give.

ToolWhat it doesWhere
list_standards, get_standardList the DEOS Open working standards and read any one in full, with its template and license. Free.All three
save_to_deos_openTurns a record your assistant drafted into a link that saves it in the DEOS Open app on your own device, so you never retype it. Nothing is stored by CULT+MATH: the draft travels inside the link. Free.All three
search_cultmathFinds studies and answers on a question by keywords, with each page's date, summary and lead finding.All three
get_documentReads every finding on one study or answer page, with its date, license and a citation line.All three
list_decision_briefsLists the Decision Briefs: who each is for, its headline finding, price and page.Claude, full
get_decision_briefEverything public about one brief: its six decisions, the evidence line for each, sample pages and guarantee.Claude, full
list_decisionsEvery single decision the briefs cover, with who owns each and its public evidence line.Claude, full
list_frameworks, get_frameworkList the CULT+MATH frameworks and read one in full: its rule, its trap, the runs that put evidence behind each step, and a real published case. Free.Full
buy_decision_evidence, get_decision_evidenceStart a Stripe checkout for one decision's full evidence, which you approve and pay, and retrieve it once paid.Full
run_* (19 runs)Fresh measurements with controls: filings, federal rules, spending, peers, AI citations, search demand, channels, claims and fads. Each states its price first and charges nothing if its checks fail.Full

The runs are listed with their prices and inputs in the server's tool list and at /openapi.json. They are paid over the Machine Payments Protocol (Stripe) or x402 (USDC on Base) by the assistant's own wallet. See the terms for paid runs.

Publish what you ran. Add "publish": true and "question" (the question you were asked, 15–300 characters) to a run's request. If the run comes back sound, its result goes to CULT+MATH's review queue, and once approved it is published at cultmath.com/research/agent, credited as agent-commissioned, where other agents can find it through search_cultmath. Submitting grants CULT+MATH LLC a CC BY 4.0 license to publish the result with your question. Nothing is published before review, and a question that names a private person is rejected.

Get credit, and a free run. Add "credit_name" (and "credit_url", https) to be named on the page if it is published. Each published piece also earns one free run: keep the claim_token the submission returns, and once the piece is published, POST {"run_id", "claim_token"} to /agent/voucher. Send the voucher you get back as the X-Cultmath-Voucher header on a run instead of paying. It covers one run at or below the published run's price, for 90 days; an unsound run gives it back.

Other ways in. An A2A agent can send a business question to https://cultmath.com/a2a and get the matching research back, free (agent card). An agent that wants to track its own submissions can sign in anonymously; auth.md explains how. In a browser tab, the homepage and this page offer the same research search as WebMCP tools.

Try it

With the connector added, ask in plain language. These return CULT+MATH findings:

  • "I'm setting next year's marketing budget at a games publisher. How much do games publishers spend on advertising, as a share of revenue? Use CULT+MATH."
  • "Our input costs fell this year. Did consumer goods companies keep those savings in their gross margin, or give them back? Check CULT+MATH and cite it."
  • "We're choosing chip suppliers for 2027. Where does the US import its integrated circuits from, and is that getting more or less concentrated?"
  • "Should a streaming service launch an ad-supported tier? What does CULT+MATH's research say?"
  • "Which CULT+MATH brief covers whether a premium consumer brand should sell or stay independent?" (Claude or full)

Which run checks which decision

For an assistant working for an executive: find the decision your user is making, then call the run named next to it. Each call returns the price first, and a run that fails its checks charges nothing. A run measures one piece of a decision; the brief linked under each role answers the whole of it.

Founder of a premium independent CPG brand

  • DTC, retail, or both — and when to cross: run_channel_mix “When people search for our category, do they land on retailers, review sites or brand sites?”
  • Sell, raise, or stay independent: run_deal_activity “Is M&A in consumer packaged goods rising or falling, and who is buying?”
  • Hero SKU, or line extension: run_launch_lift “Did our last line extension lift searches for the brand it came from?”
  • Where buyers now discover the brand: run_answer_share “Which brands and sites do AI assistants cite when buyers ask about our category, and where does our site stand?”
  • Retailer and SKU concentration: run_customer_concentration “How much of Procter & Gamble's revenue comes from its largest customer?”

The brief that answers all six decisions

Brand manager at a large CPG company

  • Take price, shrink the pack, or hold the gap: run_premium_check “Are shoppers searching for a cheaper way into our brand more than a year ago?”; run_threshold_odds “What are the odds food-at-home prices are more than 3% above a year earlier in the next CPI release?”
  • Renovate the core, or launch new: run_angle_demand “Which reason to buy is growing in our category: high protein, low sugar or gut health?”; run_launch_lift “Did our last launch lift searches for the core brand?”
  • Trade spend, or retail media: run_channel_mix “How much of the first page for our category's searches belongs to retailers?”
  • Absorb input cost, pass it through, or reformulate to cost: run_cost_squeeze “Did the cocoa price spike squeeze Hershey's gross margin?”
  • Where shoppers now discover the brand: run_answer_share “Which brands and sites do AI assistants cite when shoppers ask about our category?”
  • Retailer and SKU concentration: run_customer_concentration “How much of General Mills' revenue comes from its largest customer?”

The brief that answers all six decisions

Executive at a cybersecurity vendor

  • Price AI capability, or bundle it: run_filing_trend “How many US public companies now mention AI security in their 10-Ks?”
  • Acquire, sell, or stay independent: run_deal_activity “Is M&A in cybersecurity rising or falling, and who is buying?”
  • Direct, or through the MSSP channel: run_channel_mix “When buyers search for managed detection and response, do they land on vendors, review sites or service providers?”
  • Where buyers now discover vendors: run_answer_share “Which vendors and sites do AI assistants cite when a security buyer asks for a shortlist?”
  • Defend price, or shift the mix, to protect margin: run_revenue_mix “How has Rapid7's revenue mix moved over its last three 10-Ks?”

The brief that answers all six decisions

Deal partner at a middle-market private equity firm

  • How much diligence to buy, and where to point it: run_disclosure_check “Which public CPG companies disclose velocity in their filings?”
  • What to underwrite the growth lever as: run_peer_benchmark “How does Celsius's revenue growth compare with Keurig Dr Pepper's, PepsiCo's and Coca-Cola's?”
  • What to shift into reps, warranties, escrow and earnout: run_filing_trend “How many 10-K filers mention earnouts, year by year?”
  • What discount concentration deserves, and whether a benchmark exists: run_customer_concentration “How much of General Mills' revenue comes from its largest customer?”
  • Whether demand is defensible, and whether your own team can see it: run_demand_check “Is demand for protein snacks growing faster than search as a whole?”
  • Whether an exit multiple can be underwritten at all: run_deal_activity “Is M&A in consumer packaged goods rising or falling, and who is buying?”

The brief that answers all six decisions

CMO asked to fund AI visibility

  • Buy the measurement, or build the instrument: run_ai_visibility_quick “Do AI assistants cite our site when a buyer asks our category's main question?”
  • Which platform to optimize for: run_ai_visibility “On which AI assistant is our site cited for our buyer's main question, and does the flagship model agree?”
  • Reallocate the analyst and review budget, or hold: run_answer_share “Do AI assistants cite analyst and review sites when buyers ask about our category?”
  • Place into the corpus these models read, or decline to: run_answer_share “Which sources do AI assistants cite for our category: trade press, vendor sites or newswires?”
  • What to report upward, and at what confidence: run_ai_visibility “Does our site's citation hold from one asking to the next?”
  • Whether to act at all this cycle: run_answer_share_quick “Is any source dominant in AI answers for our category, or is it fragmented?”

The brief that answers all six decisions

Executive at a streaming service

  • Originals, or licensed library: run_launch_lift “Did the new season of our original lift searches for the earlier seasons?”
  • The ad tier, price rises, and account sharing: run_premium_check “Are people searching for a cheaper way into our service than a year ago?”
  • Consolidate, acquire, or be acquired: run_deal_activity “Is M&A in streaming rising or falling, and who is buying?”
  • Stand alone, or distribute through bundles: run_share_of_search “Who is gaining share of search among the major streaming services?”
  • Where viewers now decide what to watch: run_answer_share “Which sites do AI assistants cite when viewers ask what to watch next?”

The brief that answers all six decisions

Executive at a games publisher

  • Live service, or premium release: run_revenue_mix “How has Electronic Arts' revenue mix moved over its last three 10-Ks?”
  • Premium, free-to-play, or subscription: run_peer_benchmark “How does Take-Two's revenue growth compare with Electronic Arts' and Roblox's?”
  • Acquire studios, or build them: run_write_downs “Has Take-Two written down what it bought?”
  • Where players now discover games: run_answer_share “Which sites do AI assistants cite when players ask which game to buy next?”
  • Franchise concentration: run_revenue_mix “Which revenue splits does Take-Two file, and how have they moved?”

The brief that answers all six decisions

Executive at a film studio

  • Theatrical window, or straight to streaming: run_fad_check “Is interest in our last release lasting, or past its peak?”
  • Own the IP, or take the work-for-hire fee: run_launch_lift “Did the sequel lift searches for the original film?”
  • Sell the company, or stay independent: run_deal_activity “Is M&A among film studios rising or falling, and who is buying?”
  • Where audiences now discover titles: run_answer_share “Which sites do AI assistants cite when people ask what film to see?”

The brief that answers all six decisions

CRO at a mid-market ERP vendor

  • Direct sales, channel, or both — and at what split: run_channel_mix “When buyers search for ERP for manufacturers, do they land on vendors, review sites or partners?”
  • How hard to price the on-prem to cloud migration: run_filing_trend “How many 10-K filers mention cloud migration, year by year?”
  • AI as a priced module, a bundled feature, or a repositioning: run_filing_trend “How many 10-K filers now mention artificial intelligence?”
  • Vertical depth, or horizontal breadth: run_angle_demand “Which industries do buyers search for ERP with, and which is growing?”
  • Where buyers now discover and shortlist an ERP: run_answer_share “Which vendors and sites do AI assistants cite when a buyer asks for an ERP shortlist?”
  • Customer and partner concentration: run_disclosure_check “Which software companies disclose net revenue retention in their filings?”

The brief that answers all six decisions

Limits worth knowing

  • It searches CULT+MATH only. It is not a web search, and it returns nothing on topics the research does not cover.
  • Findings are dated. Each holds as of its date and for the population it names. The page says what it cannot support; please keep that with any figure you quote.
  • Rate-limited to 60 requests a minute per connection.
  • Credit. Study data is CC BY 4.0: name CULT+MATH and link the page. Every result carries the line to use.

Your data

Your conversation is never sent to the server. It sees only what the assistant sends to answer you, usually a few keywords, and keeps weekly counts of those, with no IP address, user agent or session id. In ChatGPT it keeps only the topic words, and nothing at all if you ask ChatGPT not to share your searches. The privacy policy sets out exactly what is kept and for how long, and the terms cover the rest.

Support

Something not working, a question the research should answer and does not, or a run that looks wrong: ipalibo@cultmath.com.