# `LangExtract.Template`
[🔗](https://github.com/mdepolli/lang_extract/blob/v0.10.0/lib/lang_extract/template.ex#L1)

The extraction task definition: a plain-language description of what to
extract, plus few-shot examples showing the expected output.

There is no schema DSL and no fine-tuned model — the template is rendered
into a few-shot prompt by `LangExtract.Prompt.Builder`, and the examples
are what teach the model the class vocabulary, span granularity,
attributes, and output format.

Build templates with `LangExtract.template!/2`, which accepts plain maps
for examples and validates them against the production aligner;
`LangExtract.template/2` is the non-raising twin for runtime task
definitions (user-uploaded or JSON-loaded). The struct is public for
pattern matching and introspection.

# `t`

```elixir
@type t() :: %LangExtract.Template{
  description: String.t(),
  examples: [LangExtract.Template.Example.t()]
}
```

---

*Consult [api-reference.md](api-reference.md) for complete listing*