quantustik

Use Quantustik from ChatGPT

Custom GPT Actions against a public OpenAPI schema, or an MCP connector where your plan supports one.

Quantustik is a risk-first market-intelligence layer for humans and AI agents, delivering calibrated probabilistic forecasts, transparent track records, and machine-readable decision support through REST and MCP. Everything on the two Claude pages in this series runs the same way in ChatGPT; only the connection step differs.

Disclaimer. Everything below is educational: how to point an AI client at Quantustik's own data and read the result. Nothing here is investment advice, a solicitation to buy or sell any security, or a personalized recommendation. Signals and forecasts are probabilistic model outputs with explicit uncertainty, not promises. Write your own prompts to ask for data, analysis, and explanation, never "tell me what to buy." Full text: Disclaimer & Terms of Use.

1. Pick a connection method

If your ChatGPT plan supports adding a remote MCP connector, point it at the same endpoint used for Claude: https://quantustik.com/mcp (streamable-HTTP, no API key required to start). If it does not, build a Custom GPT instead, which works on any ChatGPT plan that can create GPTs.

2. Build a Custom GPT with Actions

  1. In ChatGPT, go to Explore → Create a GPT → Configure.
  2. Under Actions, choose Import from URL and paste the public OpenAPI schema: https://quantustik.com/api/v1/openapi.json.
  3. Leave Authentication set to None to use the anonymous quota, or set it to API Key / Bearer if you have grabbed a free key at quantustik.com/developers.
  4. Save. The GPT can now call every endpoint described in that schema, including market conditions and per-ticker forecasts.

3. Ask a first question

What's the current market conditions read, and what's the calibrated 3-month forecast for MSFT? Give me the probability of an up move and the confidence band, not a recommendation.
Illustrative example. The GPT calls GET /api/v1/market and GET /api/v1/forecast/MSFT and might answer: “Market conditions score NN (Neutral). MSFT 3-month forecast: NN% probability of an up move, 90% confidence band −X% to +Y%.” The values above are placeholders; the real figures move daily, so ask it yourself for the current read.

4. Repeat it on a schedule

ChatGPT's Tasks feature can run a saved prompt on a recurring schedule and message you the result.

Every weekday at 8am: check the market conditions endpoint and the 3-month forecast for AAPL, MSFT, and NVDA, and send me a short plain-English summary. Do not tell me what to buy.

Congress trades, same note as the Claude routine

Congressional trade disclosures are live over REST (https://quantustik.com/api/v1/congress/feed), but that route is kept out of the public OpenAPI schema for now, so a Custom GPT built from Actions cannot call it, and it is not an MCP tool either. If your GPT has browsing enabled, ask it to fetch that URL directly; otherwise check quantustik.com/congress yourself.

More ways to run this

Disclaimer. Everything below is educational: how to point an AI client at Quantustik's own data and read the result. Nothing here is investment advice, a solicitation to buy or sell any security, or a personalized recommendation. Signals and forecasts are probabilistic model outputs with explicit uncertainty, not promises. Write your own prompts to ask for data, analysis, and explanation, never "tell me what to buy." Full text: Disclaimer & Terms of Use.