> ## Documentation Index > Fetch the complete documentation index at: /people-search/docs/llms.txt > Use this file to discover all available pages before exploring further. # Enrich person > Enrich a LinkedIn or Reddit user profile. Send linkedin_url for existing LinkedIn workflows, or profile_url for either platform. Reddit enrichment returns the public profile/avatar and one page each of recent posts and comment previews, with separate cursors and independent history errors. Professional contact and social-matching options apply only to LinkedIn. Pass a Reddit or LinkedIn `profile_url`, or use the existing `linkedin_url` field for LinkedIn. Send exactly one URL field. ## LinkedIn Pass a `linkedin_url` from [Search people](/people-search/docs/api-reference/people-search.md), or one you already have. The person's full public LinkedIn profile is collected from AnyAPI first; a 404 means no profile was found for that address. Send `platforms: []` to get only the profile. Contact lookup is off by default. With `contact: true`, `person.emails` and `person.phones` hold the person's work and personal emails and phone numbers from Crustdata, looked up alongside the platforms. Each email has a `type`, `work` or `personal`, and a `status`: `deliverable`; `catch_all`, when the domain accepts any address so delivery isn't certain; or `unknown`. Addresses Crustdata found to bounce are left out. The lookup costs Crustdata credits, about 1 plus 1 for work emails and 2 each for personal emails and phone numbers found, and fails independently: if the lookup fails, the contact lists are empty and the rest of the response still returns. `person.posts` holds the person's `linkedin_posts` most recent LinkedIn posts (10 by default, up to 100), newest first, read from LinkedIn alongside the platforms. Reposts are included and labeled by `type`: `post` is the person's own; `repost` is someone else's post shared without comment, so its `text` is empty, its `url` is that post's and `referenced` holds that post's author and text; `quote` is someone else's post shared with the person's own `text`, with the quoted post in `referenced`. `created_at` is the day the person posted or shared it. One read usually returns up to 100 posts and costs the same however many; asking for more than 50 sometimes takes a second read. A person with fewer posts than asked for returns fewer. If the lookup fails, `person.posts` is empty and the rest of the response still returns. Send `linkedin_posts: 0` to skip it. Each platform runs in parallel and fails independently: if one platform's search fails, its result list is empty and the others still return. Accounts and videos are filtered to confident matches when ranking succeeds. See [Matching](/people-search/docs/concepts/matching.md). With `youtube` in `platforms` (the default), `videos` lists YouTube videos from searching the person's name and, when their profile has a current employer, the name with that employer. Each video is checked for whether the person speaks in it, as a speaker, guest, interviewee or panelist, on any channel, and only matches come back. Videos where others talk about them or their work are left out. ## Facebook Facebook is included in the default platforms. To request it alone, send: ```json theme={null} { "linkedin_url": "https://www.linkedin.com/in/jane-doe", "platforms": ["facebook"], "linkedin_posts": 0 } ``` `accounts.facebook` contains confident matches found by looking up the person’s name and comparing public Facebook profile information with LinkedIn. `max_results.facebook` controls how many candidates are checked (1–7, default 7). A matching name or verification badge alone is insufficient. Unreadable profiles and ambiguous candidates are omitted; if the comparison fails, no Facebook matches are returned. Personal profiles and the person’s own public creator Pages are eligible; employer and fan Pages are not. This flow uses AnyAPI’s Facebook endpoints, without People Data Labs. With `enrich_accounts: true`, each match includes its profile and up to `account_posts` available public posts, reading at most five provider pages. Private or unavailable content is not accessible. Post failures preserve the matched profile and any posts already read, with `enrichment.posts_error` explaining the failure. `account_posts: 0` skips posts. Facebook following lists are not fetched. Direct Facebook URLs are not accepted as the input to Enrich person; start with LinkedIn. ## Matches come back enriched In the same call, as soon as each platform's accounts are ranked, every matched account is enriched: its `enrichment` holds the account's profile, its `account_posts` most recent posts and the `account_following` accounts it follows. Up to 8 matched videos get their `transcript`: the captions' whole text, as a string. These are billed with this call, and fail on their own: an account that can't be enriched has `enrichment: { "error": "…" }`, and a video without captions has `transcript: null` and a `transcript_error`. Send `enrich_accounts: false` or `transcribe_videos: false` to skip them. ## Reddit Send `profile_url: "https://www.reddit.com/user/username/"`, or click **Enrich** on a Reddit search result. Enrichment reads that account directly; it does not infer a real name, match other social accounts, or look up contact details. `person` contains `platform: "reddit"`, the username, public display name, avatar (`picture_url`), bio (`summary`), account creation date, karma split, public activity counts, followers, trophies, and available account flags. Missing values are null or empty. The provider's profile counts may differ from the activity returned by its history sources. A missing or suspended profile returns 404. By default, the response includes one whole provider page in each of `reddit.posts` and `reddit.comments`, newest first. Pages vary in size; this is recent public activity, not a guarantee of complete account history. Each feed contains `items` and `next_cursor`. Send `reddit_posts: false` or `reddit_comments: false` to skip a feed; disable both for just the profile. Comments contain `body_preview`, `is_preview: true`, the parent post title, and `thread_url`. The text may be shortened and the link opens the thread, not the individual comment. Posts include their title, available body text, date, subreddit, and canonical Reddit link. To continue, pass a feed's `next_cursor` back for the same account. Disable the other feed so it is not fetched again: ```json theme={null} { "profile_url": "https://www.reddit.com/user/username/", "reddit_posts": false, "reddit_comments": true, "reddit_comments_cursor": "" } ``` The dashboard provides **Load more** buttons for both feeds. A null cursor means the source supplied no continuation. We retain whole pages so following a cursor does not skip items hidden by a result limit. History lookups fail independently: the profile and any successful feed remain available. A failed feed has an `error`, an empty `items` list, and its requested cursor for retry. This is different from a successful empty history. Each request reads the profile again and bills actual provider receipts, including paid history failures. No model call is needed. LinkedIn-specific contact, platform, and video options do not trigger extra lookups for Reddit. ## OpenAPI ````yaml POST /v1/people/enrich openapi: 3.1.0 info: title: Constants Social Search API version: 1.0.0 description: >- Search for people with a natural-language description, enrich a person's LinkedIn profile with their X, Instagram, TikTok and Facebook accounts and YouTube videos, enrich any one account with its profile and recent posts, and read the transcripts of a person's videos. servers: - url: https://www.constants.io/people-search description: Production security: - apiKey: [] paths: /v1/people/enrich: post: summary: Enrich person description: >- Enrich a LinkedIn or Reddit user profile. Send linkedin_url for existing LinkedIn workflows, or profile_url for either platform. Reddit enrichment returns the public profile/avatar and one page each of recent posts and comment previews, with separate cursors and independent history errors. Professional contact and social-matching options apply only to LinkedIn. operationId: enrichPerson requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/PersonEnrichRequest' examples: every: summary: Every platform value: linkedin_url: https://www.linkedin.com/in/satyanadella some: summary: Some platforms value: linkedin_url: https://www.linkedin.com/in/satyanadella platforms: - x - youtube max_results: x: 10 youtube: 20 responses: '200': description: >- The person's profile, each platform's matched accounts and, with `youtube`, matched videos, with the matched ones enriched and transcribed. content: application/json: schema: $ref: '#/components/schemas/PersonEnrichResponse' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '404': $ref: '#/components/responses/ProfileNotFound' '413': $ref: '#/components/responses/TooLarge' '429': $ref: '#/components/responses/RateLimited' '500': $ref: '#/components/responses/Failed' '502': $ref: '#/components/responses/ProfileProviderFailed' '503': $ref: '#/components/responses/Unconfigured' components: schemas: PersonEnrichRequest: type: object properties: linkedin_url: type: string maxLength: 500 description: >- The person's LinkedIn profile address, such as `https://www.linkedin.com/in/jane-doe`. Country subdomains and addresses without `https://` work too. platforms: type: array items: $ref: '#/components/schemas/Source' minItems: 0 default: - x - instagram - tiktok - facebook - youtube description: >- Platforms to search. `youtube` returns videos in `videos` rather than accounts. Empty returns only the profile. Facebook looks up the name, reads public profiles and returns only confident matches; failed comparisons return no Facebook matches. contact: type: boolean default: false description: >- Look up the person's emails and phone numbers on Crustdata, alongside the platforms. Costs Crustdata credits: about 1, plus 1 for work emails and 2 each for personal emails and phone numbers found. linkedin_posts: type: integer minimum: 0 maximum: 100 default: 10 description: >- Up to how many of the person's most recent LinkedIn posts, reposts and quotes to return in `person.posts`, newest first. One read usually returns up to 100 and costs the same however many; asking for more than 50 sometimes takes a second read. A person with fewer posts returns fewer. 0 skips them. max_results: type: object additionalProperties: false properties: x: type: integer minimum: 1 maximum: 50 default: 50 description: X accounts to return. X bills each account returned. tiktok: type: integer minimum: 1 maximum: 30 default: 30 description: TikTok accounts to return. TikTok returns at most 30. facebook: type: integer minimum: 1 maximum: 7 default: 7 description: Facebook candidates to look up and compare, up to 7. youtube: type: integer minimum: 1 maximum: 50 default: 20 description: >- Videos to keep from each YouTube search (the name, and the name with the current employer). Above 20 uses a slower search. description: >- Results per platform. Leave a platform out for its default. Instagram's search takes no count. enrich_accounts: type: boolean default: true description: >- Enrich each matched account (its profile, recent posts and who it follows) into its `enrichment`. Billed with this call. Nothing is enriched on a platform where ranking didn't run. account_posts: type: integer minimum: 0 maximum: 100 default: 20 description: >- Recent posts per enriched account. Private accounts return none. Facebook reads up to 5 provider pages; fewer posts may be available. account_following: type: integer minimum: 0 maximum: 500 default: 200 description: >- How many accounts each enriched account follows to return. Instagram's come with only their name, picture and badges. 0 skips it. Facebook following lists are not fetched. transcribe_videos: type: boolean default: true description: >- Add the transcript of up to 8 matched videos as plain text, the captions' whole text. Billed with this call. Nothing is transcribed when ranking didn't run. profile_url: type: string maxLength: 500 description: >- A Reddit or LinkedIn user profile URL. Send exactly one of profile_url or linkedin_url. reddit_posts: type: boolean default: true description: >- For Reddit profiles, include one whole page of recent posts. False skips it. reddit_comments: type: boolean default: true description: >- For Reddit profiles, include one whole page of recent comment previews. False skips it. reddit_posts_cursor: type: string maxLength: 8000 description: >- reddit.posts.next_cursor from the previous response for this account. Keep reddit_posts enabled. reddit_comments_cursor: type: string maxLength: 8000 description: >- reddit.comments.next_cursor from the previous response for this account. Keep reddit_comments enabled. oneOf: - required: - linkedin_url - required: - profile_url PersonEnrichResponse: type: object properties: person: anyOf: - $ref: '#/components/schemas/Person' - $ref: '#/components/schemas/RedditPerson' accounts: type: object description: Keyed by platform, best first. properties: x: type: array items: $ref: '#/components/schemas/RankedAccount' instagram: type: array items: $ref: '#/components/schemas/RankedAccount' tiktok: type: array items: $ref: '#/components/schemas/RankedAccount' facebook: type: array items: $ref: '#/components/schemas/RankedAccount' description: >- Confident Facebook matches from public profile evidence. Empty when evidence is insufficient or matching fails. videos: type: array items: $ref: '#/components/schemas/RankedVideo' description: >- With `youtube` in `platforms`: the videos the person speaks in, from searching the name and, when the profile has a current employer, the name with that employer. Only matches; best first. usage: $ref: '#/components/schemas/Usage' ms: type: integer reddit: $ref: '#/components/schemas/RedditActivity' Source: type: string enum: - x - instagram - tiktok - facebook - youtube description: >- What Enrich person can search: the account platforms, plus `youtube` for videos. Person: type: object description: The public LinkedIn profile collected from AnyAPI. properties: source: type: string enum: - anyapi id: type: string name: type: string headline: type: string location: type: string current_role: type: string links: type: object additionalProperties: type: string description: >- Keyed by `linkedin`, `x`, `github` or `facebook`. `linkedin` is always set. picture_url: type: - string - 'null' emails: type: array description: >- Work emails first, then personal, up to 10. Empty unless `contact` was on and Crustdata had some. Addresses Crustdata found to bounce are left out. items: $ref: '#/components/schemas/Email' phones: type: array items: type: string description: Phone numbers as Crustdata gives them, up to 5. posts: type: array items: $ref: '#/components/schemas/LinkedinPost' description: >- With `linkedin_posts` above 0: the person's recent LinkedIn posts, reposts and quotes, newest first. Empty when the lookup failed or found none. summary: type: string connections: type: - integer - 'null' followers: type: - integer - 'null' experience: type: array items: $ref: '#/components/schemas/Experience' description: Newest first, up to 30. education: type: array items: $ref: '#/components/schemas/Education' skills: type: array items: type: string description: Inferred by the profile provider; can be noisy. languages: type: array items: type: object properties: language: type: string proficiency: type: string RedditPerson: type: object required: - platform - username description: >- Public Reddit account metadata. Name is u/; display_name is an unverified public display title. Null means the provider did not supply the field. No real-world identity or contact matching is performed. properties: source: type: string enum: - anyapi platform: type: string enum: - reddit id: type: string provider_id: type: string username: type: string name: type: string display_name: type: string picture_url: type: - string - 'null' summary: type: string links: type: object properties: reddit: type: string created_at: type: - string - 'null' karma: type: - number - 'null' post_karma: type: - number - 'null' comment_karma: type: - number - 'null' posts_count: type: - number - 'null' comments_count: type: - number - 'null' followers: type: - number - 'null' trophies: type: - number - 'null' is_employee: type: - boolean - 'null' is_nsfw: type: - boolean - 'null' verified: type: - boolean - 'null' RankedAccount: allOf: - $ref: '#/components/schemas/Account' - type: object properties: linked: type: boolean description: >- Enrich person only: true on the X account the person's professional profile links to, which is returned instead of a name search when it's found. Absent on other accounts. enrichment: description: >- Enrich person only, on matched accounts when `enrich_accounts` is on: the account's profile, posts and following, or `{ error }` when enriching it failed. oneOf: - $ref: '#/components/schemas/AccountEnrichment' - type: object properties: error: type: string required: - error RankedVideo: allOf: - $ref: '#/components/schemas/Video' - type: object properties: transcript: type: - string - 'null' description: >- Enrich person only, on up to 8 matched videos when `transcribe_videos` is on: the video's whole caption text as plain text. Captions don't say who is speaking. Null when the video has no captions or they couldn't be read. transcript_error: type: string description: Why `transcript` is null. Usage: type: object description: What the call cost. properties: credits: type: integer description: The credits the call was charged. credits_remaining: type: integer description: >- The workspace's credits left after this call. Calls running at the same time can change it. Left out when the balance couldn't be read; the x-credits-remaining header has the same number. RedditActivity: type: object properties: posts: type: object description: >- Present when requested. One whole provider page; page sizes vary. Errors do not discard the profile or other history. properties: items: type: array items: $ref: '#/components/schemas/RedditPost' next_cursor: type: - string - 'null' description: >- Opaque cursor for this feed and account. Null means this source supplied no continuation. On failure the requested cursor is retained for retry. error: type: string description: >- Present only if this history lookup failed; an empty list with error does not mean no activity. comments: type: object description: >- Present when requested. One whole provider page; page sizes vary. Errors do not discard the profile or other history. properties: items: type: array items: $ref: '#/components/schemas/RedditCommentPreview' next_cursor: type: - string - 'null' description: >- Opaque cursor for this feed and account. Null means this source supplied no continuation. On failure the requested cursor is retained for retry. error: type: string description: >- Present only if this history lookup failed; an empty list with error does not mean no activity. Error: type: object properties: error: type: object properties: message: type: string Email: type: object properties: email: type: string type: type: string enum: - work - personal status: type: - string - 'null' enum: - deliverable - catch_all - unknown - null description: >- Whether Crustdata found the address accepts mail: `deliverable`; `catch_all`, when the domain accepts any address so delivery isn't certain; or `unknown`. LinkedinPost: type: object description: A post from the person's LinkedIn activity. properties: url: type: - string - 'null' created_at: type: - string - 'null' format: date description: The day the person posted or shared it, in UTC. type: type: string enum: - post - repost - quote description: >- `post`: the person's own. `repost`: someone else's post shared without comment; `text` is empty, and `referenced` and `url` are that post's. `quote`: someone else's post shared with the person's `text`. text: type: string has_video: type: boolean media_urls: type: array items: type: string description: Images attached to the post. referenced: type: object description: 'For a repost or quote: the post it shares.' properties: author: type: - string - 'null' author_url: type: - string - 'null' description: The author's LinkedIn profile address. text: type: - string - 'null' url: type: - string - 'null' media_urls: type: array items: type: string unavailable: type: boolean description: True when a quote's original didn't come back. metrics: type: object properties: reactions: type: - integer - 'null' comments: type: - integer - 'null' shares: type: - integer - 'null' Experience: type: object properties: title: type: string company: type: string location: type: string start: type: string examples: - October 2024 end: type: string description: Empty while current. months: type: - integer - 'null' current: type: boolean description: type: string Education: type: object properties: school: type: string degree: type: string start: type: - integer - 'null' end: type: - integer - 'null' Account: type: object properties: platform: type: string enum: - x - instagram - tiktok - facebook id: type: string handle: type: string name: type: string url: type: string format: uri avatar_url: type: string bio: type: string location: type: string followers: type: - integer - 'null' following: type: - integer - 'null' posts: type: - integer - 'null' description: Posts, or videos on TikTok. verified: type: boolean private: type: - boolean - 'null' description: >- Whether the account is private, when known. Facebook does not report this. created_at: type: - string - 'null' format: date-time category: type: string description: Public account category or verification label, when available. links: type: array items: type: string description: Links from the profile, up to 5. AccountEnrichment: type: object properties: account: $ref: '#/components/schemas/Account' posts: type: array items: $ref: '#/components/schemas/Post' following: type: array items: $ref: '#/components/schemas/Account' description: >- Present when `account_following` was above 0 and the account isn't private. Each followed account leaves out `avatar_url` and `platform` (always the enriched account's), leaves out `url` and `id` when it has a `handle` (the address is the platform's profile address for that handle), and leaves out every field that came back empty, null or false, so many of them stay short. Absent for Facebook. ms: type: integer posts_error: type: string description: >- Facebook posts could not be fully loaded. Any posts already fetched remain in posts; the matched profile is retained. Video: type: object properties: platform: type: string enum: - youtube id: type: string description: The YouTube video ID. url: type: string title: type: string channel: type: string channel_url: type: - string - 'null' published: type: string description: As YouTube shows it, such as `2 years ago`. views: type: - integer - 'null' duration: type: string description: Such as `41:07`. description: type: string description: The snippet YouTube shows in search results. thumbnail_url: type: string RedditPost: type: object properties: id: type: string author: type: string created_at: type: - string - 'null' subreddit: type: string votes: type: - number - 'null' title: type: string text: type: string url: type: - string - 'null' comments_count: type: - number - 'null' upvote_ratio: type: - number - 'null' is_locked: type: - boolean - 'null' is_archived: type: - boolean - 'null' media: type: array items: type: object properties: type: type: string url: type: string RedditCommentPreview: type: object description: >- User-history comment text may be truncated. thread_url opens its parent thread; it is not a direct comment permalink. properties: id: type: string author: type: string created_at: type: - string - 'null' subreddit: type: string votes: type: - number - 'null' body_preview: type: string is_preview: type: boolean const: true post_id: type: string post_title: type: string thread_url: type: - string - 'null' Post: type: object description: A social post identified by its URL. Post IDs are not included. properties: url: type: - string - 'null' created_at: type: - string - 'null' format: date-time type: type: string description: >- X: `post`, `reply`, `quote` or `repost`. Instagram: `photo`, `video`, `carousel` or `reel`. TikTok: `video` or `photos`. Facebook: `post`. text: type: string lang: type: - string - 'null' description: X only. referenced: type: object description: 'X only: the post a reply, quote or repost refers to.' properties: author: type: - string - 'null' text: type: - string - 'null' url: type: - string - 'null' unavailable: type: boolean description: True when the original was deleted or is private. metrics: type: object properties: likes: type: - integer - 'null' comments: type: - integer - 'null' replies: type: - integer - 'null' reposts: type: - integer - 'null' shares: type: - integer - 'null' views: type: - integer - 'null' responses: BadRequest: description: The request is invalid. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: message: url must be an X, Instagram or TikTok profile address. Unauthorized: description: No API key was sent, or the key isn't valid. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: message: >- Send your API key as Authorization: Bearer . Sign in to the dashboard to make one. ProfileNotFound: description: No public profile was found for the LinkedIn address. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: message: No profile was found for that LinkedIn address. TooLarge: description: The request body is over 64 KB. content: application/json: schema: $ref: '#/components/schemas/Error' RateLimited: description: The profile provider is rate limiting requests. Retry shortly. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: message: LinkedIn profiles are rate limited right now; try again shortly. Failed: description: The request failed unexpectedly. content: application/json: schema: $ref: '#/components/schemas/Error' example: error: message: The request failed. ProfileProviderFailed: description: >- The LinkedIn profile provider failed or refused the request, for example because its wallet is out of funds. content: application/json: schema: $ref: '#/components/schemas/Error' Unconfigured: description: The server is missing the credentials this request needs. content: application/json: schema: $ref: '#/components/schemas/Error' securitySchemes: apiKey: type: http scheme: bearer description: >- An API key from the dashboard's API Keys page (sign in or sign up there with a constants.io account), sent as `Authorization: Bearer `. Keys start with `cps_`. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.