Extract - Lev8 API

POST /v1/extract routes the request by type and returns social lists, GitHub lists, or structured lists from web pages. Use Extract when you need follower, following, stargazer, or contributor lists, or when a public web page contains list-like data that should be returned as structured JSON.

Headers

Header	Required	Description
`Content-Type: application/json`	Yes	Other media types return `422`.
`Accept: application/json`	Recommended	Declares that the client expects a JSON response.
`Authorization`	Yes	`Bearer lev8_live_...` — an active Lev8 API key with Extract access.
`Idempotency-Key`	No	Stable key for safely replaying the same logical request. Lev8 generates and returns one when omitted.

Request body

Field	Type	Required	Default	Description
`type`	string enum	Yes	-	Extraction type. See supported extraction types.
`target`	string	Yes	-	Account handle, GitHub repository, repository URL, or web page URL, up to 2,048 characters.
`instruction`	string	For `web2list`	-	Describes the list items and fields to extract from a web page, up to 4,000 characters.
`limit`	integer	No	`100`	Maximum number of social or GitHub list results, from 1 to 10,000. `web2list` normalizes this field to `100`.

Supported extraction types

`type`	Target format	Example	`instruction`
`x_followers`	X username	`sama`	Not required
`x_following`	X username	`sama`	Not required
`github_user_followers`	GitHub username	`torvalds`	Not required
`github_user_following`	GitHub username	`torvalds`	Not required
`github_repo_stargazers`	`owner/repo` or repo URL	`openai/openai-python`	Not required
`github_repo_contributors`	`owner/repo` or repo URL	`openai/openai-python`	Not required
`ins_followers`	Instagram username	`openai`	Not required
`ins_following`	Instagram username	`openai`	Not required
`tiktok_followers`	TikTok username	`openai`	Not required
`tiktok_following`	TikTok username	`openai`	Not required
`web2list`	Web page URL	`https://www.forbes.com/30-under-30/2026/ai`	Required

Response

The response includes success, data, and metadata. Social, GitHub, and web2list results return data as an array. Business-level failures return success: false and a failure reason in data.

{
  "success": true,
  "data": [
    {
      "name": "Example Person",
      "company": "Example Co"
    }
  ],
  "metadata": {
    "type": "web2list",
    "target": "https://www.forbes.com/30-under-30/2026/ai",
    "instruction": "Extract name and company",
    "limit": 100
  }
}

Errors outside the Extract workflow use the standard error envelope.