AI Blocks › Hire and look after your people
Onboarding document reader
Give it an onboarding pack. Get identity, ownership, source of funds and a document list in set fields.
Onboarding document reader is a free AI skill for the job "Hire and look after your people". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · Apache 2.0 licence · Made by nighomni123 · Thinking: low
Made by nighomni123. See the original on GitHub; we copied it as it is, with its licence.
What it does
Raw client or investor onboarding papers turned into one neat record, so your team can check the facts instead of hunting for them. Nothing gets guessed. If a field is not in the documents, it stays empty. The skill first lists every document received, by type: identity, company formation, ownership, address, source of funds and tax forms. Then it pulls out names, dates of birth, addresses, owners and controllers, where the money comes from, and any tax details, into set fields. It treats the documents as untrusted text. It reads data out of them and ignores any instruction written inside them. The result feeds straight into a rules check step.
You give it
- The onboarding documents
- The type of client: person, company or trust
- The fields your firm needs (optional)
- Your document checklist (optional)
You get
- A document inventory
- A structured record of identity and ownership
- Source of funds details
- Empty fields where nothing was found
Use it when
- A new investor sent a packet of twenty PDFs and you need the key facts in one place.
- You must list which documents you have and which are missing.
- Someone retypes passport and company details into a form by hand today.
- You want a consistent record before running compliance checks.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can read your files.