We use cookies to operate this site, measure performance, and improve your experience. See our Privacy Policy or manage your privacy choices.

    Demo kits / Financial Services

    Financial Services demo kit

    The lending book of Coppermill Bank, a fictional mid-market commercial bank watching one borrower deteriorate, plus a real legal-entity corpus from the GLEIF LEI system, packaged as a loadable knowledge graph with eighteen competency questions and their expected answers. It loads into a free Fluree ledger in about two minutes.

    Create a Free Workspace

    No signup needed. After downloading, hand it to a coding agent, load it in a free workspace, or run it locally: the setup section covers all three, and takes about two minutes either way. Background reading: Fluree for Financial Services.

    Kit facts

    Download
    21 KB zip, 8 files
    Model
    19 classes and 66 properties
    Data
    171 entities, 75 real GLEIF records
    Questions
    18, each with an expected answer
    Standards used
    FIBO Business Entities · ISO 17442 LEI
    Data source
    CC0, GLEIF LEI data
    Runs on
    Free hosted workspace, or the Fluree CLI
    Updated
    August 2026
    The Scenario

    Coppermill Bank, August 2026

    Two layers in one graph: a fictional lending book with a working credit story, and a real registry corpus typed against FIBO. The questions join what core banking, risk, documents, and KYC normally keep apart.

    Coppermill lends to five borrower groups across freight, food, metals, packaging, and healthcare: $94.5M committed, $59.7M outstanding. Keystone Freight Group, its largest exposure, breached its leverage covenant at the June 30 test. The warnings were on file months earlier: a late compliance certificate, a CFO call note, a sector bulletin. Every borrower, person, and number is synthetic.

    Beside the fiction sits a real corpus: 75 legal-entity records pulled from the GLEIF LEI API, public domain under CC0, typed against FIBO’s business-entity ontologies and carrying per-record provenance. Fictional credit facts attach only to fictional borrowers; no credit property can even point at the real records.

    The flagship module is policy: the same questions asked as the bank’s credit analyst and as an external auditor return two governed answers from one ledger.

    What you can answer with it

    • Full exposure a borrower’s position across facilities, covenants, and ratings
    • Covenant headroom catch the breach while there is still room to act
    • Early warnings which signals were on file, and how early
    • Decision trail tie a credit decision to the evidence behind it
    • Audit access answer auditors without exposing pricing or personal data

    The documented trail

    1. Feb 20Q4 compliance certificate arrives six days late, a first
    2. Apr 9Call note: CFO flags volume softness, asks about headroom
    3. May 14Sector bulletin: truckload rates down 18% year over year
    4. Jun 30Q2 test: leverage 3.8x against a 3.5x cap. Breach
    5. Jul 24Waiver with conditions: fee, weekly forecasts, capex limit
    6. Aug 21Credit Committee downgrades Keystone from 5 to 7
    Setup

    Loading the kit

    Three ways to run the kit, all free and all about two minutes. Every step is also written out in the README inside the download.

    Fastest · Your coding agent does it

    Paste this into Claude Code or Cursor

    Any agent with terminal access can install the CLI, load the kit, validate it, and then answer the packaged questions and check itself against the expected answers that ship inside.

    I downloaded finserv-demo-kit.zip from flur.ee. Please set it up and check that it works:
    
    1. Unzip it and read README.md.
    2. Install the Fluree CLI if I don't have it: brew install fluree/tap/fluree
    3. Create a ledger and load the files in the order the README gives.
    4. Run the SHACL validation and confirm the data conforms.
    5. Then open competency-questions.md. Answer each question by querying the graph,
       and compare your answer to the expected answer written under it.
    
    Tell me which questions matched, which did not, and what the graph let you
    answer that separate systems could not.
    Or · No install

    Fluree AI, hosted

    A workspace in the browser with the AI agent built in. Free tier, no credit card. This is the path for asking the questions in plain English.

    Or · Local, for developers

    The Fluree CLI

    A single binary installed with Homebrew. This is the path for running the included SPARQL queries directly and for scripting.

    brew install fluree/tap/fluree

    CLI reference and install docs at docs.flur.ee

    1. 01

      Create an empty ledger

      In the workspace, add a new ledger; on the CLI, run fluree create. A ledger is a Fluree database, empty until you load it.

    2. 02

      Import the five data files

      Load model.jsonld first, then the four instances files. The model file defines the vocabulary the instance files use.

    3. 03

      Run the questions

      Ask the AI agent the competency questions as written and compare each response against the expected answer. CLI users can run the included SPARQL queries directly.

    finserv-demo · load and validate
    $ fluree create finserv-demo
    $ fluree insert -f model.jsonld
    Committed t=1 · 527 flakes
    $ fluree insert -f instances-lei-corpus.jsonld
    Committed t=5 · 980 flakes
    $ fluree validate --shacl shapes.json
    Conforms: true (6 shapes checked)
    Competency Questions

    Questions with the answers attached

    Eighteen questions ship in the kit, each with the expected answer, so you can verify the responses are grounded rather than guessed. Here is one, exactly as packaged.

    CQ3

    What early signals preceded the Q2 breach, and how much lead time did they offer?

    Expected answer

    The Q4 compliance certificate arrived six days late on February 20, the first late certificate in the relationship (130 days before the breach test); the CFO flagged shipper volume softness on an April 9 call (82 days); a sector bulletin reported spot truckload rates down 18% year over year on May 14 (47 days). The signals sat in documents; the covenant numbers still passed until June.

    The twelve credit questions

    Each is answerable only by combining more than one source domain. The note under each describes what the query touches.

    1. 01Where do we stand on Keystone Freight Group overall?Aggregates three facilities and joins the latest risk assessment
    2. 02Which covenant did Keystone breach in Q2, by how much, and what happened next?Joins the test, its covenant limit, and the waiver action
    3. 03What early signals preceded the Q2 breach, and how much lead time did they offer?Compares document dates against the June 30 test date
    4. 04What conditions did the July waiver attach?Reads the conditions recorded on the credit action, verbatim
    5. 05Which facilities mature within the next 12 months, and what is committed there?Filters maturities across the whole book
    6. 06How exposed is the book to the freight sector?Connects borrower industry to committed exposure and the index

    Six further questions ship in the kit: three over the real GLEIF corpus and three over the policy module.

    Modules

    Extend the kit, one capability at a time

    Each module is a separate download that adds one capability to the same scenario. Load the core kit first; each module’s README carries the full recipe with its outputs pasted verbatim.

    Module · 3 questions

    Governed access

    One question, asked by different people, returns different answers, enforced by the database rather than by prompts or separate copies.

    Give an external auditor the covenant and rating trail they need while client pricing, document contents, and beneficial owners stay out of reach.

    One insert adds two identity records and five stored access policies. Pass the identity with --as; the query text never changes.

    Download the module (4 KB, 5 files)

    q-facilities.sparql · same query, twice
    $ fluree query -f q-facilities.sparql --as …#creditAnalyst
    8 rows · spreadBps populated on every row
    $ fluree query -f q-facilities.sparql --as …#externalAuditor
    8 rows · spreadBps blank · owners return 0 rows
    Downloads

    What is in each zip

    The core kit is eight files. The policy module is a separate zip, self-contained with its own README. Nothing requires a signup.

    finserv-demo-kit.zip

    Download (21 KB)
    • model.jsonldThe vocabulary: FIBO entity typing plus the fin: credit extension. Annotated throughout.
    • instances-book.jsonldThe bank, five borrower groups, six borrowers, eight facilities, covenants, guarantees.
    • instances-risk.jsonldCovenant tests, ratings, credit actions with cited evidence, the freight index, the document trail.
    • instances-kyc.jsonldBeneficial owners and completed KYC reviews. Fictional persons.
    • instances-lei-corpus.jsonld75 real GLEIF records, per-record provenance.
    • shapes.jsonSHACL integrity rules, run in-database with fluree validate.
    • competency-questions.mdAll 18 questions with expected answers.
    • README.mdLoad instructions, standards references, data-source and license details.
    finserv-policy-module.zip4 KB · 5 files

    Governed access

    policy.jsonld, three queries, README with verified outputs

    Download

    The shipped data conforms to the included SHACL shapes; break it and fluree validate names the violations.

    Support

    Questions about the kit

    For a guided walkthrough of the kit, or to talk about running the same pattern on your own data, contact us.