> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.rightbrain.ai/v-1/docs/concepts/input-processors/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.rightbrain.ai/_mcp/server.

# Input processors

> Attach URL fetching, document extraction, audio transcription, or web search to a Task revision and inspect each processor's configuration contract.

An **input processor** transforms one input before the Task prompt is rendered. Processors belong to a Task revision, so changing them creates a new revision and leaves previous revisions unchanged.

Each entry in `input_processors` contains:

| Field             | Contract                                                 |
| ----------------- | -------------------------------------------------------- |
| `param_name`      | A `{placeholder}` present in the revision's user prompt. |
| `input_processor` | A processor name returned by `GET .../input_processor`.  |
| `config`          | Optional processor-specific settings.                    |

Revision validation rejects a `param_name` that is absent from the user prompt and more than one `audio_transcriber`. Processor names and configurations must match the registered processor contract before processing can complete.

## Available processors

| Processor                    | Input                                           | Result                                              |
| ---------------------------- | ----------------------------------------------- | --------------------------------------------------- |
| `url_fetcher`                | URL string                                      | Replaces the URL with fetched page content.         |
| `document_content_extractor` | Uploaded document                               | Replaces the target parameter with extracted text.  |
| `audio_transcriber`          | Uploaded WAV, MP3, AIFF, AAC, OGG, or FLAC file | Replaces the target parameter with a transcription. |
| `perplexity_search`          | Search query string                             | Replaces the query with ranked web results.         |

The catalog endpoint returns a plain array with each processor's name, description, and `config_schema`. Use the returned schema instead of assuming that all processors accept the same settings.

`document_content_extractor` currently requires the uploaded file's name in `config.filename`, although its catalog entry does not publish a configuration schema. Only one `audio_transcriber` can be configured on a revision.

The processed value replaces the target parameter before prompt rendering. Large fetched pages, search results, transcripts, or extracted documents therefore increase the Task model's input context. Track `input_processor_timing`, tokens, and credits before increasing source size or retrieval breadth.

## Files and direct model input

Document extraction and audio transcription consume files supplied with a Task run. File-capable models can instead receive supported image or PDF files directly through the revision's `file_input_mode`; an input processor is only needed when the file must be converted before the model runs.

Run records expose `input_processor_timing`. Processor charges, when present, are reported separately in `input_processor_charged_credits` and included in the run's total charged credits.

## Related

#### [Tasks](/docs/concepts/tasks)

Configure the prompts, files, and output contract that use processed inputs.

#### [Run tasks via the API](/docs/api/run-tasks)

Send structured and multipart Task inputs.

#### [API Reference: Input Processors](/api-reference/api-reference/tasks)

Inspect the live processor catalog and revision schemas.

> Build reliable AI agents that run inside your existing tools and workflows. Rightbrain developer documentation.