Build the canonical source of truth for a new client in under 10 minutes. Every proper noun, every team member, every tool — documented once, used by every agent.
Paste whatever you have from onboarding — notes, intake forms, emails, LinkedIn profiles, session transcripts. The skill extracts every proper noun, team member, and tool, organizes them into a standard reference data structure, and flags the names that audio transcription is most likely to mangle.
Not a template. A builder — one that catches the proper nouns you’d forget to include until an agent misspells them in a client deliverable.
Each section of the reference data document handles a different part of client identity.
Works with any large language model. Setup steps vary slightly by platform — most support custom instructions or uploaded skill files.
Start a new conversation. Type “Build reference data” or “Set up a new client” and paste whatever you have — onboarding notes, intake forms, emails, LinkedIn profiles, session transcripts.
The more sources you paste, the more complete the reference data. Intake forms plus emails plus LinkedIn profiles together catch more than any single source.
One file. Upload it to your AI assistant and you’re set.
Download Skill FileWorks with any large language model.