FACE — Framework for AI Context in Enterprise

Your AI tools don't know your business.
FACE fixes that.

Every session your team re-explains who you are, what your tone is, what your standards are. FACE loads that context automatically — into every AI tool, for every person, every time.

Open FACE KB Manager IT Admin guide →

Works with ChatGPT, Claude, Copilot, Gemini, Slack, Teams & more.  ·  First knowledge base free, forever.

The problem

Your team is doing AI's job for it — every day.

This isn't an AI problem. It's a context problem. The models are capable. They just don't know your business.

Output that sounds like it was written for someone else's company
Staff copy-pasting the same briefing into every prompt, every session
Inconsistent AI output across your team — same question, different answers

What is FACE?

A company briefing for every AI tool you use — loaded automatically.

Build one structured knowledge base — your brand, processes, standards, team. Every AI tool your team opens already knows all of it. No briefing. No copy-paste. No generic output.

Three layers. Always the right context, always the right person.

Layer 1 — Company

Everyone

Brand voice, mission, standard processes, company-wide policies. Loaded for every person, every session.

Layer 2 — Department

By role

Marketing sees marketing context. Sales sees sales context. Each team gets exactly what they need — nothing more.

Layer 3 — Product / Project

By team

Project-specific knowledge, automatically loaded when relevant. Scoped to the people who need it.

Your data stays in your own private repositories. FACE and KB Manager never store copies of your content. The source code is publicly auditable — so you can verify that promise yourself.

How it works

Set up once. Works for everyone, automatically.

Everyday users do nothing different. They open ChatGPT or Claude as usual — the context is already there.

You build the knowledge base

Your brand, your processes, your tone — structured in FACE format using KB Manager. ABQ Institute can build it for you, or your IT team can set it up in a few hours.

KB Manager connects to every AI tool

One connection. Every AI tool your team uses — ChatGPT, Claude, Copilot, Gemini, Slack agents, Teams bots — reads from the same knowledge base automatically.

Your team gets better output — immediately

No setup for your staff. No learning curve. The AI responses are on-brand, accurate, and consistent — for everyone on your team, from day one.

Anyone who wants to use AI on their own terms

The model is rented.
The agent is yours.

Skills, memory, context — all stay with you. Better AI shouldn't mean starting over. New model. Same agent. Nothing lost.

Upgrade the intelligence. Keep everything you've built.

Who it's for

If your team uses AI — FACE makes it work properly.

Accounting & Finance

Your AI knows your client categories, billing formats, and compliance language.

Stop copy-pasting policy documents into every prompt. Client emails come out right the first time, every time.

Marketing & Creative

Every piece of copy reflects your brand voice — not a generic AI template.

Briefs, proposals, campaign copy: all contextualised to your agency and your client. Consistent across your whole team.

Solo Operators & Consultants

Your AI works like a well-briefed assistant — one that never forgets.

Your pitch, your service offering, your proposal format — loaded automatically. Stop starting from zero every session.

Small & Medium Businesses

Consistent AI output for everyone — whether they started last week or three years ago.

No knowledge locked in one person's head. Everyone's AI works from the same playbook.

KB Manager

Five steps. Then your whole team is running on FACE.

KB Manager is the app that connects your knowledge base to every AI tool. Get started at face.abq.institute.

GitHub account

Free to create at github.com

Log in to KB Manager

face.abq.institute

Create your company

Connect your GitHub org

Connect your LLM

ChatGPT, Claude, Gemini…

Create & populate the KB

Your team is live

Pricing — pay per knowledge base

Pricing scales with your organisational complexity — not your headcount. Each additional knowledge repository (department layer, product layer) is billed at a small flat monthly rate.

Repositories Per repo / month Monthly total Typical use
1 FREE €0 Solo operator, small team, pilot
2–3 €19 €19–€38 Company + 1–2 departments
4–5 €15 €53–€68 Company + 3–4 departments
6–7 €12 €80–€92 Multi-dept + product teams
8–10 €10 €102–€122 Full organisation rollout
11+ Enterprise — contact ABQ  ·  unlimited repos, SSO, audit logs, SLA

€0 / month

1 repository — Free forever

Each repository = one structured FACE knowledge base (Company, Department, or Product layer). The first repository is free forever — no credit card required. You upgrade only when your organisation grows into additional layers.

Questions

No. If your company has set up FACE, you use your existing tools exactly as before — Slack, Teams, ChatGPT, whatever you already use. The technical layer is invisible to everyday users.
ChatGPT, Claude, GitHub Copilot, Cursor, Gemini, Slack agents, and Microsoft Teams bots — and any AI tool your organisation adopts in the future. The knowledge base is tool-agnostic.
No. ABQ Institute can handle the full setup for you, or your IT team can follow the admin guide below. Either way, everyday users need zero setup — they just open their AI tools as normal.
Tell your FACE administrator. Wrong answers usually mean a piece of context is missing or outdated. They update the knowledge base, and the AI answers correctly for everyone from that point on.
No. All knowledge lives in your own private GitHub repositories — under your organisation's full control. KB Manager reads from them; it does not store copies. You can revoke app access at any time. The KB Manager source code is publicly auditable so you can verify this independently.
You pay per knowledge repository — not per user. The first repository is free forever. Additional repositories are billed at a small monthly rate that decreases slightly as you add more layers (see the pricing table above). This reflects organisational complexity, not headcount — everyone on your team uses FACE at no extra cost.

Nine steps to a fully deployed FACE knowledge base.

Most organisations complete setup in 4–6 hours. Prerequisites: GitHub account (free), KB Manager access, and an AI coding tool (Claude Code, Cursor, or Copilot).

Phase 1 — Prepare

Step 1

Run a Friction Audit

30-min interviews with 5–10 people across roles — find where knowledge gaps slow work down. Without this, you build on assumptions. ABQ can facilitate →

Step 2

Secure leadership commitment

Appoint an AI Adoption Lead with at least 20% of their time protected. 70–80% of AI programmes fail because production pressure kills transformation work.

Phase 2 — Build

Step 3

Connect your knowledge platform

KB Manager uses GitHub by default. Already on Confluence or Slite? Connect via MCP — no migration needed. Choose your scenario:

  1. Go to github.com/organizations/plan — create a free organisation
  2. Add your IT lead and AI Adoption Lead as owners
  3. Continue to Step 4 — KB Manager will create the repository inside it
  1. In KB Manager, click "New organisation" and connect your existing GitHub org via OAuth
  2. KB Manager creates a new [org]-kb-company repository from the FACE template
  3. Your org structure is unchanged — the KB repository is the only new addition
  1. Connect your GitHub organisation in KB Manager
  2. Create a new FACE-structured repository alongside your existing ones — they stay untouched
  3. KB Manager connects to the new FACE repository only
  1. In KB Manager, select "Connect via MCP" and choose your platform
  2. Authenticate via OAuth — KB Manager connects to your designated KB space
  3. Agents propose updates as unpublished drafts for your team to review
  1. In KB Manager, click "New organisation" and connect your GitHub org
  2. Select "Import existing KB" and point to your [org]-kb-company repository
  3. KB Manager indexes it directly — nothing is created or modified

Step 4

Create your Company Knowledge Base

Skip if you chose "Import existing KB" in Step 3.

  1. Use the FACE KB Company Template
  2. Name it [your-org]-kb-company — set to Private
  3. Clone locally, open in your AI tool, run:
    Read AGENTS_SETUP.md and guide me…

Delete AGENTS_SETUP.md after setup. AGENTS_BOOT.md stays.

Step 5

Create your first Product Knowledge Base

  1. Same template — name it [your-org]-kb-prod-[product]
  2. Follow the same setup as Step 4
  3. Populate: architecture decisions, runbooks, team structure, roadmap

Step 6

Invite users and assign permissions

RoleGitHub permissionWhat they can do
AdminOrg ownerFull access, manage teams and repositories
Context ChampionWrite accessPropose and edit KB content via pull requests
Regular userRead accessLoad context into AI tools, use Slack/Teams agents
AI agentRead accessLoad context automatically, propose updates for human review

When someone leaves: remove from GitHub org — all access revoked automatically.

Phase 3 — Launch

Step 7

Connect KB Manager

  1. Log into KB Manager
  2. Connect your GitHub org (OAuth — ~2 min)
  3. Select repositories to index as knowledge layers
  4. Done — context loads automatically for every user

Roadmap: M365 SSO — permissions mapped from M365 groups.

Step 8

Run the pilot — 4 weeks

Measure: Are people using AI with context daily? Have "explain this" prompts dropped? Is output on-brand and immediately usable?

Output: Metrics, before/after examples, recommendation to expand.

Step 9

Expand and scale

Once the pilot proves value, follow the Implementation Roadmap. Context champions handle ongoing maintenance. Quarterly reviews keep the KB current.

Developer & admin FAQ

No. Use your existing GitHub organisation. Create new FACE-structured repositories inside it — your existing repos are untouched.
Yes — if your team already uses Confluence or Slite, connect via MCP (Model Context Protocol). No migration required. GitHub remains the recommended default for new setups. Microsoft 365 SSO is on the roadmap.
GitHub repository permissions are the access boundary. If someone cannot read a repository, they cannot load it as context. When someone leaves, removing them from the GitHub organisation revokes all access automatically.
KB Manager's core is published under the Functional Source Licence (FSL-1.1). You can run it on your own infrastructure for free for internal use. Offering it as a hosted service to third parties requires a commercial licence from ABQ Institute. Each released version automatically converts to Apache 2.0 two years after release.

Open source & licensing

Built in the open. Auditable by anyone.

The source code is public — so any organisation can verify the security promises FACE makes, not just read them.

FACE Framework

GNU LGPL v3.0

Use FACE internally without open-sourcing your own knowledge layers. If you modify the framework mechanisms, those modifications return to the commons. Copyright: ABQ Institute.

KB Manager App

FSL-1.1 → Apache 2.0

Source code is public and auditable. Free for internal use — self-host at no cost. Commercial hosting requires a licence from ABQ. Converts automatically to Apache 2.0 two years after each version release.

Full licence texts are in the GitHub repositories. For enterprise legal review or commercial licence questions, contact ABQ Institute.