PeriodFinder in ChatGPT
Use PeriodFinder inside ChatGPT and Claude
The cross-brand absorbency and sizing tools, answered live inside your AI assistant. Ask in plain language and get the same verified, graded numbers the site runs on, with links back to the products and the full table.
What you can ask
Once connected, just ask in a normal chat. For example:
- “What Knix tier equals Thinx Super, in millilitres?”
- “I need about 45 mL of coverage. What tier should I buy in each brand?”
- “What size am I in every brand if my hip is 40 inches?”
- “Show me in-stock period underwear that covers a heavy day, cheapest first.”
- “How much does "Heavy" actually hold, brand by brand?”
Behind those answers are four tools: translate any brand tier to millilitres and to every other brand, size across brands from one hip measurement, find in-stock products that cover a target absorbency, and pull the citable dataset itself.
How to connect it
In ChatGPT, turn on Developer Mode (Settings, then Connectors, then Advanced), start a new chat, open the + menu, add a connector, and paste this address:
https://www.periodfinder.com/api/mcp/mcpIn Claude or another MCP client, add the same address as a custom connector (Streamable HTTP). It is public and read-only, so there is nothing to sign in to.
The same verified data as the site
Every answer comes from PeriodFinder's cross-brand absorbency table, where each brand's own claim is normalized to millilitres and graded for source quality. See the full graded table, the methodology, or the machine-readable dataset.
Questions
- Is the PeriodFinder ChatGPT app free?
- Yes. It answers from the same public data the website runs on. There is no account and no charge to use it.
- What data does it collect?
- None about you. The tools take a measurement or a brand and tier as input and return an answer; nothing is stored, and no health information is written to any URL. It is the same privacy-by-design approach as the site.
- Does it work in Claude and other assistants too?
- Yes. It is a standard MCP server, so any MCP client (Claude and others, in addition to ChatGPT) can add the same endpoint as a connector.
- Where do the numbers come from?
- From each brand's own published claims, normalized to millilitres and graded A/B/C for source quality, the same verified data documented on the methodology page. The app links back to the full graded table so you can check any figure.