> ## Documentation Index
> Fetch the complete documentation index at: https://deepline.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# deepline describe

> Describe anything search returns: a tool, monitor type, or play.

## describe

```text theme={"languages":{"custom":["/languages/deepline-prompt.json"]}}
Usage: deepline describe [options] <name>

Describe anything search returns: a tool, monitor type, or play.

Options:
  --prebuilt                Show the Deepline tool or play, not a play your
                            workspace created under the same name
  --tool                    Same as --prebuilt
  --verbose                 Show secondary fields, and fill in every input and
                            read every output in the examples
  --cli-example             Show only the CLI example, and explain how it takes
                            inputs and returns outputs
  --play-example            Show only the in-play example, and explain how it
                            takes inputs and returns outputs
  --no-example              Hide the run examples
  --full-schema             Print every field: a long schema unfolded (over 100
                            lines), list rows, and a tool's raw response
  --fields <names>          Show only these fields, comma-separated, plus every
                            required input
  --feature <name[=value]>  Describe one feature of a tool, such as a model:
                            --feature model=openai/gpt-5.6-luna. A tool lists
                            its features.
  -o, --out <path>          Save the output to this file instead of printing it
  --schema-only             Print only the schema, with every field, and nothing
                            else
  --json                    Print JSON instead of text; with --schema-only, only
                            the input and output JSON Schemas
  --web-url                 Print only a play's play page URL
  -h, --help                display help for command
```

```text theme={"languages":{"custom":["/languages/deepline-prompt.json"]}}
Explanation:
  Purpose:
    The contract for anything search returns: a tool, a play, or a monitor
    type. What it is, cost, inputs, outputs, and an example of running it from
    the CLI and from inside a play.
  When:
    After search and before the first run of something you have not used. You
    do not say which kind the id is.
    Exact tool ids can also resolve supported tools hidden from list/search
    for compatibility. Absence from discovery does not mean a tool was removed.
  Input:
    The id or reference from search. --verbose fills in every input and reads
    every output in the examples; --cli-example and --play-example print one
    example; --schema-only prints only the schema; --full-schema unfolds a
    schema over 100 lines; --out <path> saves the output; --json prints JSON.
    A tool lists its features, such as the model an AI tool runs;
    --feature <name>[=<value>] describes one.
    A play your workspace created is shown instead of a Deepline tool or
    prebuilt play of the same name; deepline/<name> shows the Deepline one.
  Output:
    Readable text by default; read that form. With --json:
    kind, id, displayName, description, cost, inputs[], outputs[], and
    runExamples {cli, play}, plus the kind's own fields.
  Cost:
    Free.
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.