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

# Medical-device guidance change review

> Track changes between dated regulator guidance versions, such as FDA medical device guidance, in a cited change table with old and new passages.

**Example question:** What changed between the FDA’s 2005 and 2023 guidance on software documentation in premarket submissions?

**You get:** A change table and CSV with old and new passages, section references, document dates and applicability questions.

**Start with:** Authority and jurisdiction; topic; older and newer document dates or URLs; product context for the reviewer.

## Run with DeepResearch

Set `VALYU_API_KEY` in your environment, then run this illustrative request. Change the question and CSV fields for your own task. The CSV fields below describe the requested output, not a saved result.

<Accordion title="Run the research request">
  ```bash cURL theme={null}
  curl --fail-with-body https://api.valyu.ai/v1/deepresearch/tasks \
    -H "x-api-key: $VALYU_API_KEY" \
    -H "Content-Type: application/json" \
    --data-binary @- <<'JSON'
  {
    "query": "Compare the FDA guidance Content of Premarket Submissions for Device Software Functions issued in 2023 with the 2005 Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices. Locate official FDA versions and confirm their status and dates. Focus on documentation categories and requested software documentation. Return a source-linked change table with old and new section references. If a version cannot be accessed, mark that comparison unavailable. Do not select a submission pathway, determine product compliance or infer a statutory requirement from a guidance recommendation.",
    "mode": "standard",
    "research_strategy": "Use FDA documents and official supersession notices. Exclude vendor commentary from the evidence rows. Record the publication date separately from access date. Cite every row; mark missing or inaccessible evidence.",
    "report_format": "A concise evidence table, comparison limits and unresolved questions.",
    "deliverables": [
      {
        "type": "csv",
        "description": "One row per evidence item with the requested fields and supporting source URL.",
        "columns": [
          "change_topic",
          "older_document",
          "older_date",
          "older_section",
          "newer_document",
          "newer_date",
          "newer_section",
          "change_type",
          "source_passages",
          "older_url",
          "newer_url",
          "applicability_question"
        ]
      }
    ]
  }
  JSON
  ```
</Accordion>

Save the returned `deepresearch_id`. Poll [Get status](/api-reference/endpoint/deepresearch-status) until the task completes, then read the report and download the CSV from `deliverables`.

## How it works

1. Find the official versions and their issue, draft/final and supersession information. Preserve each version as a separate record.

2. Read the relevant sections in both versions. Match requirements by meaning and section reference rather than heading alone.

3. Record additions, removals and revised recommendations with both source passages. Keep applicability questions for expert review.

To build the workflow inside your own agent, use [Search](/search/quickstart) to find sources and [Contents](/guides/content-extraction) to read selected pages. Keep the CSV fields above as your extraction schema.

## Check the result

Open both cited versions and verify that each claimed change is present. Check scope, draft/final status and later updates before applying a historical comparison to current work.


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