How to Use Schematron V2 Small in VS Code
Schematron V2 Small turns HTML into typed JSON, with particular strength on complex schemas and long pages. Choose it when extraction quality matters most; Schematron V2 Turbo prioritizes throughput.
Prompts to try with Schematron V2 Small in VS Code
Copy an example into VS Code, or ask in your own words.
Use Schematron V2 Small on the project's saved product HTML. Define a Zod schema for name, specifications, and availability, then compare results with the expected fixtures.
Replace one product parser with Schematron V2 Small for evaluation. Preserve the existing parser and report field differences on the same saved HTML.
Build a Pydantic extraction schema for the saved job pages and run Schematron V2 Small. Describe advertised salary precisely and allow absent salary values.
Prepare the project's article HTML for Schematron V2 Small by removing scripts and styles. Extract headline, author, publication date, and article link.
Use Schematron V2 Small to extract records from the saved directory HTML. Split sections so each requested JSON result stays below 4K tokens.
How to set up Schematron V2 Small in VS Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
Use Add to VS Code on the console's Agents & MCP page, or run:
code --add-mcp '{"name":"spiral","type":"http","url":"https://api.spiral.ai/mcp"}' - Sign in
VS Code asks you to sign in when it first starts the server: sign in to Spiral and click Authorize. No API key. Run MCP: List Servers from the Command Palette to check that Spiral is running.
Use the Spiral skill
- Send VS Code this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Schematron V2 Small
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "inference-net/schematron-v2-small", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: VS Code finds Schematron V2 Small on Spiral, shows the price and runs it.
Schematron V2 Small providers and prices
Spiral runs Schematron V2 Small through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.001228 per run at the default inputs. $0.065 per 1M input tokens, $0.299 per 1M output tokens.
Settings you can ask for
Ask VS Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 8192 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Inference.net: Schematron V2 Small is on its tool page: open its card above.
What can you do with Schematron V2 Small in VS Code?
Turn saved HTML into typed records your project can use. Have the coding agent define schemas, prepare page content, and check extracted fields against existing fixtures.
Product catalog ingestion
Extract product names, current sale prices, specifications, and availability from product-page HTML into consistent records.
Job posting collection
Turn job-page HTML into records containing employer, role, location, requirements, and advertised compensation. Allow missing compensation to remain null.
Article metadata indexing
Extract headlines, authors, publication dates, and article links from supplied HTML for a searchable content index.
Company directory enrichment
Convert company-page HTML into records of names, descriptions, locations, and contact details explicitly present on each page.
Recurring product monitoring
Have an agent fetch tracked pages, extract current product details, and compare the resulting records with earlier observations.
Parser replacement evaluation
Run extraction against saved HTML alongside an existing parser. Compare individual fields before replacing parsing rules.
More on Schematron V2 Small and VS Code
Overview
Schematron V2 Small in other agents
Questions about Schematron V2 Small in VS Code
Can VS Code use Schematron V2 Small?
Yes. Add Spiral to VS Code as an MCP server, and VS Code can find and run Schematron V2 Small like any built-in tool.
How much does Schematron V2 Small cost in VS Code?
Up to $0.001228 per run at the default inputs. $0.065 per 1M input tokens, $0.299 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your VS Code plan is separate.
Do I need an OpenRouter account to use Schematron V2 Small in VS Code?
No. Spiral runs Schematron V2 Small on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Schematron V2 Small in VS Code?
About two minutes: add the hosted server, then sign in.
Does VS Code wait while Schematron V2 Small runs?
Schematron V2 Small runs in the background. VS Code starts the run, checks on it and gives you the text, json when it's ready.
What is Schematron V2 Small best used for?
Extracting typed records from HTML, especially when pages are long or the desired schema is complex. Product catalogs, job postings, and article metadata are useful starting points.
How should I write extraction instructions?
Put the instructions into schema descriptions. Specify which value to extract when several appear, such as the current sale price versus the original price; conversational directions do not steer extraction.
Can Schematron V2 Small open a webpage itself?
It needs HTML supplied by your agent or crawler. It does not fetch web addresses or render JavaScript, so your agent must obtain the page content before extraction.
How much HTML and JSON can it handle?
The maker lists a 128K-token context window and a 4K-token maximum output. Split large extraction jobs into smaller sections when the requested JSON would exceed that output allowance.
What improves extraction accuracy?
Use temperature zero, remove scripts and styles, and keep the HTML that contains the desired information. Add precise schema descriptions wherever multiple values could fit the same field.
Start using Schematron V2 Small in VS Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into VS Code:
set up https://spiral.ai/SKILL.md


