1
Scope your project with Clara
Sign in to Clarifeye and click Create a new knowledge store. Clara runs a short scoping interview, around 20 minutes, to understand what your project is about and who runs it. She then proposes the artifacts to build, drafts a first version of them along with a project plan, and shows you what needs your attention first. The more context you give her, the more useful that first version, but it is a starting point you build up from, not a finished store.
2
Bring in the context
Under Sources, upload documents and connect drives like SharePoint or Google Drive. Plug in tools such as Confluence, Granola or your IT systems as external MCPs. Clara reads everything and flags gaps and contradictions before anyone is interviewed.
3
Run interviews
Open Interviews. Clara proposes the gaps worth exploring and drafts the interviews; you assign each one to the person who holds that knowledge. Clara runs them asynchronously, in the contributor’s own language, with every question informed by what is already known. See Interviews.
4
Consolidate
As interviews and sources come in, go to Work with Clara. Clara reads new input against the current artifacts, surfaces contradictions and gaps, and proposes concrete edits to your artifacts, from the mental map to the playbooks. You review and steer every change, and she proposes the next step. See Work with Clara.
5
Put the knowledge to work
How you use a store depends on the use case.
- The store is the deliverable. On a project like an ERP migration or an AI deployment, the knowledge is what you hand over: the integrator, the consultants or the team building an agent take it as the specification of how you actually work. They read it in Clarifeye or pull it into their own tools through the Clarifeye skill. See Connect to AI clients.
- The store is the tool. When the goal is to answer questions on a policy or produce a recurring deliverable, use the store directly. Three surfaces sit at the same level: Chat and Task inside Clarifeye under Put knowledge to work, and MCP through the Clarifeye skill in Claude, ChatGPT, Copilot or a custom client. See Use and Connect to AI clients.
6
Test it with Chat or Task
Under Put knowledge to work, Chat is for short exchanges: ask a question, check an answer, see which artifacts and sources it rests on. Task is for long-running jobs: hand Clara a concrete piece of work, such as filling an RFP or reviewing a contract, and inspect the output and its sources when it is done. Both apply the knowledge as it stands today. See Use.
7
Connect it to your AI client
Install the Clarifeye skill in your AI client (Claude, ChatGPT, Copilot, or a custom MCP client). From then on the client can draw on any knowledge store you have access to, applying your logic instead of generic answers. See Connect to AI clients.
8
Keep it current with signals
As people use the store, corrections and gaps surface as signals, centralized in one place. Clara helps you triage them and feed them back into the artifacts, sometimes by triggering a new interview. Encourage everyone to leave a signal whenever something looks off. See Signals.
9
Invite your colleagues
Share the knowledge store with your team so they can answer interviews, review artifacts, use Chat and Task, and raise signals. Permissions are scoped per store, so the same person can be a contributor in one store and a user in another. See User management.