
Customers, employees, and buyers reach grounded answers and product decisions by describing what they need.
Support answers arrive inside the product, so integration teams stop leaving their work to search three sources or wait on a ticket.
A platform agent that answers questions about the platform from its documentation, community threads, and knowledge base, without leaving the product.
Searches documentation, community threads, and the knowledge base in one pass, ranks what it finds by relevance and recency, and writes a sourced answer.
Teams that build and run integrations need answers all day about how the platform behaves: what a shape does, how a connector is configured, what an error means. The answer exists, but it sits in documentation, a community thread, or a knowledge base article, and reaching it means leaving the work at hand, searching three places, or raising a support ticket and waiting. The context is gone by the time the answer arrives, and a small question costs a morning.
The people asking are merchants who run their own integrations, partners and systems integrators who build for clients, and internal teams who operate the flows afterward. All of them are mid-build or mid-fix when the question lands.
The agent lives inside the product rather than beside it. It opens from the agent studio, or from a chat icon pinned in the top navigation that follows a user to any screen, so the question is asked where it arose and the answer comes back without a tab change.
A question is answered in one pass over three corpora: the product documentation, community posts and answers, and knowledge base articles. The agent searches them together, ranks the results by relevance and recency so a current page outranks an out-of-date article, and writes a single direct answer instead of a list of links, with the sources attached for anyone who wants to read further. The chat holds its context, so a follow-up can be a pronoun.
Output is prose in the chat panel plus source links, and nothing is exported: the answer stays where the work is happening. Asked how an error-handling shape behaves, the agent returns an overview of that shape, what it does, how to set it up, and how it behaves, without leaving the product.
Patchworks is behind this solution. Ask about answering integration questions inside the product, and the MACH Alliance will put you in touch with the person who can answer.
Not ready to talk to anyone? MACH Insider is the Alliance mailing list. New solutions, program openings, and the research behind them, once a month.
Everyone credited on this solution
Every organization named here has a page listing the rest of its work on the Hub.

Customers, employees, and buyers reach grounded answers and product decisions by describing what they need.
Shoppers get an answer on the product page in under a second, grounded in documents the brand controls.

Shoppers get answers about a product, in their own language, without leaving the page they are on.
Contributor submission
The detail on this page comes from the contributor's own account of their own work. The MACH Alliance reviews a submission before publishing it and does not independently verify the figures in it.
What this page does not establish
Where this came from
Solution detail is drawn from Patchworks' three-page submission to the hub, Patchworks Answers Agent, and the product documentation page it cites for the screenshots.
Put your agentic solution in front of the enterprise leaders who use the hub to decide what to build next. Submissions are open to MACH Alliance Members, Collaborators, or Ambassadors.