Guide: Vehicle Rental (SQLite)

An end-to-end canonic project on a vehicle rental service: one SQLite connection, five dimensions, three fact tables, nine metric contracts (three canonical), and one enforced guardrail. No external database required. The entire dataset lives in a single rental.db file.

Full source: examples/rental/

Schema

DIMENSIONS
  vehicle_categories   category_id · name · daily_rate
  locations            location_id · code · city · country · airport_code
  employees             employee_id · name · title · location_id
  customers             customer_id · name · email · country · membership_tier
  vehicles              vehicle_id · make · model · year · category_id

FACTS
  rentals    rental_id · customer_id · vehicle_id · pickup/dropoff_location
             employee_id · dates · planned_days · actual_days
             rate_per_day · total_amount · status
  payments   payment_id · rental_id · payment_date · amount · method · status
  damages    damage_id · rental_id · vehicle_id · severity · repair_cost

Seed data: 40 rentals (32 completed, 3 active, 2 confirmed, 2 cancelled, 1 no-show), 32 settled payments, 8 damage claims, across 20 customers, 15 vehicles, 5 locations.

Setup

sqlite3 rental.db < setup.sql   # create the database (one-time)
cd examples/rental              # canonic commands must run from here
canonic status
# Canonic project: rental-demo (version 1)

Quickstart

canonic ingest --bootstrap                 # bootstrap the semantic layer from the live schema
canonic query --metrics rental_count --dimensions status
canonic mcp start                          # start the MCP server
canonic eval baseline \                    # (optional) grain-inference accuracy
  --candidates candidates.yaml \
  --dataset eval/grain_cases.jsonl

canonic status, canonic ingest --bootstrap, canonic query, and canonic mcp start never call the LLM: every table here has a declared primary key, so grain is inferred deterministically. CANONIC_LLM_API_KEY only matters for canonic eval baseline, which explicitly benchmarks the configured model against eval/grain_cases.jsonl (a labeled set with primary keys withheld, to force drafting).

Metrics

contracts/metrics/ ships 9 metric contracts. Three are the canonical showcase bindings. The rest support the guardrail’s caveats and other reporting cuts.

Metric	Source · measure	Canonical for
`rental_revenue`	`payments.total_paid`	Settled revenue across all payment methods
`rental_count`	`rentals.completed_rental_count`	Number of completed agreements
`avg_rental_duration`	`rentals.avg_rental_days`	Mean actual days held per completed rental
`active_customers`	distinct count on `rentals.customer_id`	Customers with a non-cancelled rental
`median_rental_amount`	percentile: `rentals.total_amount`	Median order value
`ending_inventory`	semi-additive: `vehicle_inventory.inventory_level`	Fleet size as of the latest snapshot date
`damage_count` / `total_repair_cost` / `avg_repair_costs`	`damages.*`	Damage-claim reporting

Guardrail

contracts/guardrails/ ships one enforced guardrail, completed-rentals-only: rentals.total_base_revenue must never be summed without a status = 'completed' filter: active and cancelled rows have NULL in total_amount. Use payments.total_paid for financial reporting instead.

Cross-fact fanout (documented in a knowledge caveat, not a guardrail: none of the four guardrail kinds cover “reject a join between two fanning fact tables”, see Contracts & guardrails): joining damages to payments in a single query fans out both facts. Always bridge each independently through rentals.

Files

canonic.yaml                    ← SQLite connection, LLM, reconcile settings
setup.sql                       ← DDL + seed data
semantics/rental_db/            ← 5 dimensions + 4 facts (incl. vehicle_inventory)
contracts/metrics/              ← 9 metric contracts (see Metrics above)
contracts/guardrails/           ← completed-rentals-only.yaml
knowledge/global/               ← 5 pages: rental-revenue, rental-duration, active-customers,
                                   avg-repair-costs, ending-inventory definitions/notes
eval/grain_cases.jsonl          ← 8 labeled grain-inference cases (PK omitted)

Getting Started

Concepts

CLI Reference

MCP / Agent Integration

Guides

Reference

Guide: Vehicle Rental (SQLite)

Schema

Setup

Quickstart

Metrics

Guardrail

Files

​Schema

​Setup

​Quickstart

​Metrics

​Guardrail

​Files

Schema

Setup

Quickstart

Metrics

Guardrail

Files