Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build database-powered applications with Baserow, the open-source no-code database. Use when a user asks to create spreadsheet databases, build API-connected workflows, manage relational data, or self-host Baserow as an Airtable alternative.
.claude/skills/terminalskills-baserow/SKILL.md| Model | Eval pass | Runs |
|---|---|---|
| gemini-3.6-flash | 100% | 7 |
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 105% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 40% | 0% |
You are an expert in Baserow, the open-source no-code database platform and Airtable alternative. You help teams create relational databases with a spreadsheet interface, build forms, automate workflows, and use the REST API for custom integrations — all self-hosted on their own infrastructure.
bash# Docker Compose (production-ready) docker compose up -d # UI at http://localhost:80 # Or one-liner for testing docker run -p 80:80 -v baserow_data:/baserow/data baserow/baserow:latest
markdown## Table Types and Fields ### Field Types - Text, Long text, Number, Boolean, Date, URL, Email, Phone - Single select, Multiple select (colored tags) - Link to table (relationships between tables) - Lookup (pull data from linked records) - Rollup (aggregate linked records: SUM, COUNT, AVG) - Formula (computed fields using other fields) - File (attachments) - Created by, Last modified, Auto-number ### Formulas concat(field('First Name'), ' ', field('Last Name')) if(field('Status') = 'Paid', field('Amount'), 0) datetime_format(field('Created'), 'YYYY-MM-DD') year(now()) - year(field('Birth Date'))
bash# List rows with filtering curl "https://baserow.example.com/api/database/rows/table/TABLE_ID/?user_field_names=true&filter__Status__equal=Active&order_by=-Created" \ -H "Authorization: Token YOUR_TOKEN" # Create row curl -X POST "https://baserow.example.com/api/database/rows/table/TABLE_ID/?user_field_names=true" \ -H "Authorization: Token YOUR_TOKEN" \ -H "Content-Type: application/json" \ -d '{"Name": "New Project", "Status": "Active", "Priority": "High"}' # Update row curl -X PATCH "https://baserow.example.com/api/database/rows/table/TABLE_ID/ROW_ID/?user_field_names=true" \ -H "Authorization: Token YOUR_TOKEN" \ -d '{"Status": "Completed"}' # Webhooks — trigger on row events # Configure in UI: Settings → Webhooks # Events: rows.created, rows.updated, rows.deleted
Example 1: User asks to set up baserow
User: "Help me set up baserow for my project"
The agent should:
Example 2: User asks to build a feature with baserow
User: "Create a dashboard using baserow"
The agent should:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,607 | 5,416 | -53% | 1 | 1 | 0% | 1,461 | 2,005 | +37% | 0 | 0 | — |
case-02 | fail→fail | 8,857 | 4,716 | -47% | 1 | 1 | 0% | 1,725 | 1,876 | +9% | 0 | 0 | — |
case-03 | pass→pass | 5,057 | 3,565 | -30% | 1 | 1 | 0% | 888 | 1,608 | +81% | 0 | 0 | — |
case-04 | pass→pass | 8,211 | 5,369 | -35% | 1 | 1 | 0% | 1,597 | 1,879 | +18% | 0 | 0 | — |
case-05 | fail→pass | 6,779 | 3,066 | -55% | 1 | 1 | 0% | 1,322 | 1,422 | +8% | 0 | 0 | — |
case-06 | fail→pass | 4,535 | 3,320 | -27% | 1 | 1 | 0% | 731 | 1,498 | +105% | 0 | 0 | — |
case-07 | fail→pass | 6,356 | 3,677 | -42% | 1 | 1 | 0% | 1,250 | 1,515 | +21% | 0 | 0 | — |
case-08 | pass→pass | 5,646 | 3,073 | -46% | 1 | 1 | 0% | 1,019 | 1,539 | +51% | 0 | 0 | — |
case-09 | fail→pass | 6,843 | 3,967 | -42% | 1 | 1 | 0% | 1,219 | 1,708 | +40% | 0 | 0 | — |
case-10 | fail→fail | 10,197 | 4,335 | -57% | 1 | 1 | 0% | 1,722 | 1,645 | -4% | 0 | 0 | — |
case-11 | fail→pass | 9,027 | 4,802 | -47% | 1 | 1 | 0% | 1,711 | 1,785 | +4% | 0 | 0 | — |
case-12 | fail→pass | 11,005 | 7,338 | -33% | 1 | 1 | 0% | 1,851 | 2,276 | +23% | 0 | 0 | — |
case-13 | pass→pass | 10,796 | 5,670 | -47% | 1 | 1 | 0% | 2,031 | 2,007 | -1% | 0 | 0 | — |
case-14 | pass→pass | 7,997 | 4,392 | -45% | 1 | 1 | 0% | 1,339 | 1,674 | +25% | 0 | 0 | — |
case-15 | pass→pass | 14,884 | 10,337 | -31% | 1 | 1 | 0% | 2,686 | 3,055 | +14% | 0 | 0 | — |
case-16 | fail→pass | 7,520 | 3,739 | -50% | 1 | 1 | 0% | 1,239 | 1,548 | +25% | 0 | 0 | — |
case-17 | pass→pass | 13,378 | 12,455 | -7% | 1 | 1 | 0% | 2,465 | 3,340 | +35% | 0 | 0 | — |
case-18 | pass→pass | 16,701 | 11,382 | -32% | 1 | 1 | 0% | 2,877 | 2,945 | +2% | 0 | 0 | — |
case-19 | pass→pass | 8,879 | 5,309 | -40% | 1 | 1 | 0% | 1,787 | 1,826 | +2% | 0 | 0 | — |
case-20 | pass→pass | 13,912 | 11,696 | -16% | 1 | 1 | 0% | 2,678 | 3,161 | +18% | 0 | 0 | — |
case-21 | pass→fail | 4,997 | 5,141 | +3% | 1 | 1 | 0% | 1,009 | 1,955 | +94% | 0 | 0 | — |
case-22 | pass→pass | 5,849 | 3,726 | -36% | 1 | 1 | 0% | 1,152 | 1,684 | +46% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.