---
title: "Run SEO scan"
method: POST
path: "/api/apps/{app_id}/seo/scan"
---

# Run SEO scan

`POST /api/apps/{app_id}/seo/scan`

<Info>This API is in beta. Endpoints, fields, and behavior may still change, so avoid depending on it in production.</Info>

Scans the app for SEO problems and returns a score with the checks behind it.

The scan reads the app's own configuration and also fetches the live site over HTTP to check what it really serves for `robots.txt`, `sitemap.xml` and the home page. Those three run in parallel with a five-second timeout each and one retry, so expect a few seconds, and up to about ten when the live site is slow to answer. An app that isn't published yet still scans, but those live checks have nothing to fetch. What you get back for one is the gating failure itself: Base44 drops the failures that merely restate it, and collapses the unreachable-host warnings into a single row rather than repeating one per probe.

Base44 then stores the result, so [Get SEO score](/api-reference/get-seo-score) and [Get last SEO scan](/api-reference/get-last-seo-scan) serve this scan until the next one runs. That makes scanning the write in this group: prefer the two report endpoints when you only need the last result. Both writes are best-effort and can't fail the scan, so a 200 here is the scan rather than a promise about what the report endpoints will say. Read them back if you need to be sure.

The score and the checklist are stored separately, and two scans running at once can interleave those two writes. So if you read both report endpoints, compare their `scanned_at` before treating the numbers as one scan's.

Each check that Base44 can act on carries a `fix_action`. Read it to see what the builder offers; applying one isn't part of this API. `score.overall` and the per-category counts are renormalized during scoring, so they don't add up to a plain count of failing checks. Use `failures` from [Get SEO score](/api-reference/get-seo-score) for that.

This endpoint is limited to 30 requests per minute per app, and that budget is shared with the app's other SEO endpoints.

<Warning>The response includes fields beyond the ones documented here. Don't rely on undocumented response fields, as they can change at any time.</Warning>

## Path parameters

- `app_id` string, required — ID of the app to scan.

## Response `200`

The scan Base44 just ran.

- SEOScanReportResponse — A scan Base44 just ran, wrapped in the field the endpoint returns it under.
  - `result` SEOScanReport, required — An SEO scan: its score and every check behind it.
    - `score` SEOOverallScore, required — The scan's score, overall and per category.
      - `overall` integer, required — Overall SEO score for the app, 0 to 100.
      - `grade` string, required — Letter grade for `overall`: `A`, `B`, `C`, `D` or `F`.
      - `categories` object — Per-category breakdown, keyed by the same category names the checks carry: `meta_tags`, `crawlability`, `structured_data`, `ai_discoverability` and `content_quality`. Base44 can add a category, so read the map rather than assuming these five. Every category in the map is scored, including one this scan ran no checks for, which scores 100 with all three counts at zero.
    - `checks` SEOCheckResult[] — The checks behind the score, in the order Base44 reports them. Not every check it ran: a failure that only restates an upstream one is dropped in favour of the root cause, so an unpublished or login-gated app returns the one gating failure rather than the dozen findings that follow from it.
      - `id` string, required — Identifier of the check. Stable across scans, but a check can be missing from a later scan rather than change status, because Base44 collapses cascading failures into their root cause.
      - `category` string, required — Which part of the score this check feeds: `meta_tags`, `crawlability`, `structured_data`, `ai_discoverability` or `content_quality`.
      - `title` string, required — Short name of the check, phrased for the finding rather than the subject, so it changes with `status`.
      - `status` 'pass' | 'warn' | 'fail', required — Outcome of the check. A `warn` scores 50 against the category and a `fail` scores 0. Not every check can return all three.
      - `description` string, required — What the check found, written for the app's owner.
      - `details` unknown[] — Supporting detail for the finding, usually one object per page or entity involved. The keys differ per check, so treat the entries as opaque.
        - unknown
      - `fix_action` SEOFixAction — The remedy Base44 offers for a check, for display.
        - `type` string, required — Identifier of the action the Base44 builder offers for this check.
        - `label` string, required — Label the builder shows on the button for this action.
        - `params` object — Arguments to pass with the action. Empty on every action a scan emits today; keep reading it rather than assuming it stays empty.
      - `info_only` boolean — `true` when the finding is environmental, such as DNS or hosting, and can't be resolved from inside the app.
    - `scanned_at` string, required — When the scan ran, as an ISO 8601 UTC timestamp.

## Other responses

- `401` — Missing or invalid credentials.
- `403` — You don't have access to this app, or you used a workspace API key. These endpoints accept a user's credentials only.
- `404` — App not found.
- `429` — Rate limit exceeded (30 requests per minute).

---

[API](https://skmtc.dev/base44/apis/base44-app-management-api.md) · [All operations](https://skmtc.dev/base44/apis/base44-app-management-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/base44/base44-app-management-api/revisions/173e4e9c63c2/schema)