---
title: "AI overview extraction"
method: POST
path: "/ai-overview/extract"
---

# AI overview extraction

`POST /ai-overview/extract`

## Request

- Base URL: `https://www.agentseo.dev/api/v1`
- URL: `https://www.agentseo.dev/api/v1/ai-overview/extract`
- Auth: API key in header `x-api-key`

## Query parameters

- `sync` boolean

## Headers

- `x-project-id` string
- `x-workflow-id` string

## Request body

- AiOverviewParams
  - `keyword` string, required
  - `location` string
  - `location_code` integer: Optional search location code for deterministic geo-targeting.
  - `language` string
  - `target_domain` string: Optional domain to check against the sampled citation candidates.

## Response `200`

Completed AI overview workflow with agent_workflow

- AiOverviewResponse
  - `keyword` string
  - `location` string
  - `target_domain` string, nullable
  - `geo_target` GeoTarget
    - `input` string, nullable
    - `location_code` integer
    - `location_name` string, nullable
    - `country_iso_code` string, nullable
    - `source` 'resolved' | 'explicit_code'
  - `ai_overview_detected` boolean
  - `overview_status` 'detected' | 'not_detected'
  - `citations` object[]
    - `rank` integer
    - `title` string
    - `url` string, uri
  - `overview_audit` object
    - `overview_summary` string, nullable
    - `citation_domains` string[]
    - `citation_analysis` object
      - `scope` 'sampled_serp_candidates'
      - `target_domain` string, nullable
      - `target_domain_present` boolean, nullable
      - `target_domain_rank` integer, nullable
      - `competing_domains` string[]
    - `content_requirements` string[]
    - `recommended_asset_type` 'answer_page' | 'comparison_page' | 'local_landing_page' | 'guide'
    - `opportunity_level` 'high' | 'medium' | 'low'
  - `agent_workflow` AgentWorkflow
    - `summary` string
    - `decision` 'ready' | 'act_now' | 'monitor' | 'blocker'
    - `confidence` number
    - `recommended_actions` AgentWorkflowAction[]
      - `priority` 'critical' | 'high' | 'medium' | 'low'
      - `action` string
      - `why` string
    - `evidence` AgentWorkflowEvidence[]
      - `finding` string
      - `detail` string
      - `severity` 'critical' | 'high' | 'medium' | 'low'
    - `limitations` string[]
    - `next_calls` AgentWorkflowNextCall[]
      - `endpoint` string
      - `reason` string
  - `markdown_summary` string

## Other responses

- `202`: Job queued or still processing

---

[API](https://skmtc.dev/agentseo/apis/agentseo-api.md) · [All operations](https://skmtc.dev/agentseo/apis/agentseo-api/llms.txt) · [OpenAPI document](https://skmtc.dev/agentseo/apis/agentseo-api/revisions/53a62da6f92c?raw)