AI answers for food and FMCG operations.
Local by default. Hosted when you choose.

FloorMind is a governed, read-only AI query layer for food manufacturers. Ask about yield, waste, orders, allergens or batch provenance in plain English. The default Ollama setup keeps model prompts on-site. If you configure a hosted provider, the question and relevant context are sent to that provider.

๐Ÿ”’ Read-only database setup ๐Ÿ›ก 4-layer SQL safety ๐Ÿ“ Every query audit-logged ๐Ÿ  Local Ollama by default โ˜ Hosted AI optional ๐Ÿ“บ Kiosk dashboards
Why FloorMind

The reporting gap on the factory floor

Your ERP and MES systems hold the answers. Getting them out is the problem.

01

Three slow paths to one answer

Write SQL yourself (error-prone), wait for IT (slow), or export to Excel (outdated by the time it opens).

02

Plain English in, governed answer out

Anyone on the team asks in plain English. FloorMind routes the question to the right domain, generates and validates SQL, and explains the result.

03

Read-only setup, from the start

For an initial demo, use synthetic data or an approved read-only copy. Live database access starts with a DBA-configured SELECT-only account, ideally on a replica. App validation adds another layer, it does not replace database permissions.

4safety layers with a SELECT-only database setup
6business domains scoped per question
17pre-built queries answering instantly
2AI modes: on-prem by default or hosted when configured
How it works

From question to audit-logged answer in four steps

No plugins into your ERP. No data pipelines to build. One read-only connection is enough.

1

Domain routing

Your question is mapped to one of six business domains (production, traceability, orders, compliance, staff, stock) so the model only ever sees the tables that matter.

2

Fast path or LLM

17 common factory questions return pre-tested SQL without an LLM call. Other questions use the local model by default, or a hosted provider if configured.

3

Static validation

Generated queries pass sql-sop analysis and FloorMind's app validator. A SELECT-only database account remains the enforcement boundary if a query gets past those checks.

4

Execution with evidence

Results render as a table or chart with a plain-English explanation, and the full query record lands in a tamper-evident audit ledger.

See it

Real screens, synthetic data

Captured from the running demo. No mockups.

FloorMind factory dashboard: output and waste trends, delivered revenue, product yield and estimated waste cost
Factory dashboard: production trends, customer revenue, product yield and estimated waste cost in one view.
FloorMind kiosk mode: login-free dashboard for factory wall screens
Kiosk mode: login-free wall view for factory TVs. Aggregate yield, waste, and alerts only.
End-to-end examples

One question, joined data, a clear answer

Complex questions across production, sales and traceability. SQL is generated, safety-checked, executed and audit-logged.

Production performance

Which station and shift need attention?

1,075 kg scrap

VA Line night shift, across 19 runs. Conversion was 92.5%, with 7.5 minutes average downtime per run.

Delivered sales

Where is revenue strongest by retailer and SKU?

GBP 22,402

Morrisons Fish Pie Mix, from 1,428 kg delivered at a weighted GBP 15.69 per kg.

Traceability

Where did recent production runs originate?

36 runs

North Atlantic, wild-caught, Iceland. Linked to one trace ID in this sample dataset.

Plain-English question SQL generated Safety validation Result and audit record

These examples use synthetic demo data and a hosted Groq model. No customer or production data is shown.

Open the full run: questions, grounded answers and result tables
Three validated complex queries showing production performance, retailer sales and traceability results
Three LLM-generated queries against synthetic sample records. Each passed SQL validation and executed successfully.
Self-serve

Try the demo in 5 minutes

Start with synthetic data on your machine. Pre-built queries work without an API key; install Ollama for open-ended local questions. A hosted provider is optional and receives the prompts and context needed to answer.

# 1. Get the code and set up Python 3.11+
git clone https://github.com/Pawansingh3889/FloorMind.git
cd FloorMind
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt

# 2. Seed 60 days of synthetic factory data
python scripts/seed_demo_db.py

# 3. Run the app
streamlit run app.py   # โ†’ http://localhost:8501

# Optional: local LLM for open-ended questions
ollama pull gemma3:12b

# Optional: kiosk dashboard for a TV screen
FLOORMIND_KIOSK=1 streamlit run kiosk.py

Full setup guide with Mac and Linux notes: docs/setup.md

Subscription

What a subscription covers

The software runs on your premises. The subscription covers everything that keeps it working there.

Installation and connection

We deploy FloorMind on your hardware and connect it to your read-only replica: SQL Server or PostgreSQL, mapped to your schema.

Schema mapping and tuning

Your tables, your terminology, your domain rules. We tune the question router and the pre-built query library to your operation.

Support SLA and updates

Priority fixes, model and dependency updates, and a named engineer who knows your installation.

Audit-ready evidence

Every query linted, logged, and hash-chained. Governance documentation written for corporate IT and BRC auditors, not just developers.

Pricing

Flat per-site fees, in pounds

Excluding VAT. One fee per site, with no per-seat charge. Choose the support and deployment level that fits your operation.

Starter

Single site, self-serve install, evaluation and small teams.

ยฃ99/month

excl. VAT, billed monthly

  • Up to 2 concurrent users
  • SQLite or PostgreSQL demo setup
  • Pre-built query library
  • Community documentation
  • Email support, 3 business days
Start with a pilot

30-day pilot. Annual billing saves 20%.

Enterprise

Multi-site groups, custom integrations, procurement and auditor requirements.

from ยฃ1,200/month

excl. VAT, annual agreement

  • Everything in Pro, multi-site
  • Managed deployment and monitoring
  • ERP / MES integration work
  • Security review support and DPA
  • On-site pilot with success criteria
Contact us

Annual agreement, priced per deployment.

For IT and audit

Built for the security conversation

FloorMind was designed so a security reviewer can verify the controls themselves. The full model is in docs/architecture.md.

LayerControlWhat it prevents
L1Database grantsDBA-configured SELECT-only user with explicit DENY on writesAny write reaching the source database
L2Connection routingDuring setup, point FLOORMIND_DB at the approved read-only replicaAny connection to the production primary
L3Static analysisRule-based sql-sop lint before query execution; current source and package activity are linked belowUnsafe SQL patterns and poor query constructs
L4App validatorSecond independent parse with row capsDefence in depth and unbounded result sets
Open-source components

Inspect the safety stack

Open source by Pawan Singh Kapkoti, with development in the Governed Agent Stack. Source, tests and package activity are linked below; companion tools are distinguished from controls used by FloorMind.

Used in FloorMind

FloorMind SQL validator

Built into this app: read-only statement checks, injection-pattern checks, table validation, row caps and the sql-sop pass.

agent-blackbox

FloorMind's audit log can mirror events to this append-only, hash-chained ledger when installed and enabled. It records actions; it does not approve queries.

FloorMind installs it as an open-source dependency. The ledger can be verified independently.

Separate companion tools

sql-sop-mcp

Exposes the linter to MCP-compatible clients. It is not part of FloorMind's Streamlit query path.

sql-explorer-mcp

Separate read-only SQL MCP server with driver-level restrictions, SQL AST validation and sql-sop lint. Optional role checks can be added with query-warden.

query-warden

Optional role-based policy check for allowed SQL tables and columns. It is not currently wired into FloorMind.

These PePy charts include CI downloads. They show package activity, not unique users or an independent security audit. PePy version metadata can lag PyPI; use each project source or PyPI page for current release details.

See your own numbers in the demo

Start with synthetic data or an approved read-only copy on your machine. Before a production connection, your DBA configures a SELECT-only account. We can answer a question your team asks every week.