Example / Private knowledge
A private knowledge helper answers from the right documents.
An internal helper is not a folder search box wearing a chat skin. It has to know who is asking, which documents that person may see, and when an answer should become a handoff instead of a guess.
Reference workflow. Who may see what, how long it is kept, and what gets logged all have to be agreed with the organization that owns the documents.
- Works in
- ChatGPT
- Claude
- Gemini
- Coding agents
Direct answer
A private knowledge helper should check who the person is, pull only the sources they are allowed to see, name those sources in the answer, and send unresolved work to the team that owns the decision
Identity first01
Make access part of the request.
Who may see what is not an afterthought. The search request has to carry the person's identity, their team or workspace, and any role limits needed to filter the results before a single line of text goes back to the AI chat.
That keeps the rules inside your product, where someone can review and change them without rewriting the instructions given to the AI.
Model Context Protocol · authorizationAnswer with evidence02
Named sources make the answer checkable.
When an answer leans on an internal handbook, a contract, or a runbook, show the document title and the section it came from. The person can then judge whether the source fits the situation, instead of treating the answer as an unexplained authority.
It also shows which documents are slipping. An out-of-date answer points straight back at the document that needs a review.
Keep actions separate03
Knowing a policy is not permission to change a system.
The helper can explain the approved path and offer a next step. Anything that changes a system belongs behind its own named action, its own checks, and its own confirmation. A policy answer should never quietly turn into a payroll, access, or customer-record change.
That line lets a team ship useful answers first, and decide later which working actions are safe to open up.
Measure unanswered work04
Treat the gaps as your writing list.
Track the questions that found no source, turned up sources that disagreed, or needed a person. Those patterns show which documentation is missing, and which narrower, approved workflow is worth building next.
The goal is not more chat. It is a more reliable next answer and a more deliberate next action in your product.
AI / Second opinion05
Not sure where to start? Ask an AI.
Paste this question into ChatGPT, Claude, or Gemini. It walks you through finding the one task your customers would most want to finish by chatting, instead of clicking through your app.
“I run a software product. Help me identify one high-value workflow my customers could finish inside ChatGPT instead of switching between tabs. Ask me about the product, the user, the action, the data it needs, permissions, and the safest small first release.”
The prompt is copied as a backup. Some AI chats may ask you to paste it after sign-in.
FAQ06
Questions, answered.
Is this just an MCP server?
No. The server is one piece. We do the whole job: we pick the customer task, set who can do what, handle the quirks of each AI chat, make the actions run fast, add monitoring, test it, and ship it.
Do we have to rebuild our product?
Almost never. Our layer sits in front of what your product already does. We start with one small workflow and grow from what works.
Will it work in ChatGPT, Claude, and Gemini?
The same build can serve all three. Each one has its own sign-in, look, approval step, and publishing rules, so we set up and test each one separately.
How do you keep it safe?
Your product decides what is allowed, and we keep that in charge. Every action is spelled out, every input is checked, and the customer sees an approval step before anything happens. Our layer sits beside your product, not inside it.
Where does it run? Does it touch our code?
It runs as a separate service in front of your product. It never goes into your code, and you still own your data.
What does it cost, and how fast?
A first workflow starts at €3,200 and goes live in about 7 days.
Contact us07
Now build yours
Send us your product, who uses it, and the action you have in mind. We will tell you the smallest version worth building, and whether it is a €3,200 first workflow or something that needs a closer look first.