Changelog

Changed Batch enrichment — min_followers and exclude_platforms are no longer listed on the POST /public/v1/enrichment/batch/ request. Batch enrichment does not filter by follower count or platform — it enriches every row of the uploaded CSV. To work from a filtered set, run a Discovery search first and upload those creators as your CSV.

Added Has personal website filter — set has_personal_website: true under filters.creator_has on a Discovery request to filter for creators with a website on their Instagram, TikTok, or YouTube profile. Only true is meaningful; false or omitting the field has no effect.

Added Enrich by handle (Profile) and Enrich by handle (Analytics) — two lower-cost subsets of the Enrich by handle (Full) response:

• POST /public/v1/creators/enrich/handle/profile/ — identity, contact, and vetting data (validated email, name, gender, location, cross-platform presence, verification, account type, follower/media counts, bio, niche). Billed at 0.2 credits per successful request.
• POST /public/v1/creators/enrich/handle/analytics/ — performance analytics only (engagement/growth detail, posting frequency, medians, income estimates, audience data, and optional lookalike creators). Available for Instagram, YouTube, TikTok, OnlyFans, Twitter, and Twitch; the other platforms are rejected with a 400. Billed at 0.8 credits per successful request.

Added Natural-language search on Discovery — send an nlp_search brief at the top level of a Discovery request ("fitness creators in New York with over 100k followers") and the AI maps it to filters before the search runs. Filters and platform you set yourself always win. The response adds applied_filters and nlp_search so you can see exactly what ran. Billed at the normal Discovery rate.

Added Brands classifier — GET /public/v1/discovery/classifier/brands/ now documents search and paging, and page size is configurable with limit (1–100); the default is 50. Set exact=false to match part of a brand's normalized name instead of its handle — ?search=kfc&exact=false finds "KFC UK & Ireland".

Added Matched filter values (premium) — add the query parameter ?return_filter_values=true to a Discovery request to include a matched_filters object on each returned creator, echoing the filter values it matched on. Premium calls are billed at 0.03 credits per creator (this replaces the 0.01 base rate; it is not additive). Premium pricing does not change how the Discovery fair-use cap is counted.

Added Sponsored post data — Instagram Enrich by handle (Full) responses now include sponsored_posts (the creator's commercial posts) and past_sponsors (sponsors aggregated per brand handle, with post counts and first/last seen dates) when include_audience_data is enabled.

Added Discovery exclusion lists — named, platform-scoped lists of handles to exclude from Discovery results. New endpoints under /public/v1/discovery/exclusion-lists/:

Exclusion lists

• GET exclusion-lists/ — list exclusion lists
• POST exclusion-lists/ — create an exclusion list
• GET exclusion-lists/{list_id}/ — get one exclusion list
• PATCH exclusion-lists/{list_id}/ — rename an exclusion list
• DELETE exclusion-lists/{list_id}/ — delete an exclusion list
• GET exclusion-lists/{list_id}/entries/ — list handles in one list
• POST exclusion-lists/{list_id}/entries/ — add handles to a list
• DELETE exclusion-lists/{list_id}/entries/ — remove handles from a list

Default lists (per platform, auto-created on first access)

• GET exclusion-lists/default/{platform}/ — get the default exclusion list
• GET exclusion-lists/default/{platform}/entries/ — list handles in the default list
• POST exclusion-lists/default/{platform}/entries/ — add handles to the default list
• DELETE exclusion-lists/default/{platform}/entries/ — remove handles from the default list

Changed Discovery usage limit — replaced the previous separate Discovery-credit balance with a fair-use cap on unique creators. Each renewal raises the cap and anything unused carries over. Past the cap, POST /public/v1/discovery/ and POST /public/v1/discovery/creators/similar/ pause and return 429 until your subscription renews. See Usage, pricing, and limits.

Added Social listening — monitor creators and receive their activity via webhooks. New endpoints under /public/v1/social/listening/:

Subscriptions

• GET subscription/ — list subscriptions
• POST subscription/ — create a subscription
• GET subscription/{subscription_id}/ — get a subscription
• PATCH subscription/{subscription_id}/ — update a subscription
• DELETE subscription/{subscription_id}/ — delete a subscription
• GET subscription/{subscription_id}/creators/ — list subscription creators
• POST subscription/{subscription_id}/creators/ — add subscription creators
• GET subscription/{subscription_id}/details/ — subscription stats
• GET subscription/{subscription_id}/details/export/ — export stats (CSV)

Campaigns

• GET campaign/ — list campaigns
• POST campaign/ — create a campaign
• GET campaign/{campaign_id}/ — get a campaign
• PATCH campaign/{campaign_id}/ — update a campaign
• DELETE campaign/{campaign_id}/ — archive a campaign
• GET campaign/{campaign_id}/creators/ — list campaign creators
• POST campaign/{campaign_id}/creators/ — add campaign creators
• DELETE campaign/{campaign_id}/creators/ — remove campaign creators
• GET campaign/{campaign_id}/overview/ — campaign overview
• GET campaign/{campaign_id}/content/ — campaign content feed

Events

• GET events/ — list events
• GET events/{event_id}/ — get an event

Usage

• GET usage/ — current limits and active subscription counts

Added API key management — create and manage long-lived API keys under /public/v1/accounts/:

• GET tokens/ — list your API keys
• POST tokens/ — create a new API key
• PATCH tokens/{token_id}/ — update an API key's label, scopes, active status, or expiration
• DELETE tokens/{token_id}/ — permanently delete an API key

Added OAuth PKCE flow — authorization code + PKCE flow for desktop and CLI integrations (S256 only), under /public/v1/oauth/:

• POST register/ — register an OAuth client for a persistent client_id
• GET authorize/ — browser-based consent flow
• POST token/ — exchange the authorization code for an access token

Added Connected OAuth apps — manage the apps a user has authorized, under /public/v1/oauth/:

• GET apps/ — list connected apps
• DELETE apps/{app_id}/ — revoke its access

Added Similar creators — each account may include similarity_score (number): how closely that creator matches your reference (higher = stronger match).

Added Location classifier — new search, offset, and limit query parameters. Use search for substring/prefix matching (e.g. "London", "United"); paginate with offset + limit (1–100 per page).

Added Similar creators — Twitter support via the new SimilarCreatorsTwitterRequest schema. Look up by url, username, or id.

Added Enrich by handle (Full) — audience demographics in the response when include_audience_data=true. Returns breakdowns for followers, commenters, and likers: geography, languages, age, gender, interests, brand affinities, plus credibility histograms and reachability.

Added Batch enrichment — new recommended download endpoint GET /public/v1/enrichment/batch/{batch_id}/download/ returns a short-lived download URL for CSV results. The original GET /public/v1/enrichment/batch/{batch_id}/ is preserved as the direct-download fallback and is still required for JSON output via ?format=json.

Removed Discovery — exclude_previous parameter, deprecated Feb 25, 2026. Sending it now returns 400; use exclude_handles.

Added Discovery — content-exclusion filters: keywords_not_in_captions (Instagram), keywords_not_in_tweets (Twitter), keywords_not_in_video_titles (YouTube), not_link_in_bio (Instagram, TikTok, Twitter, Twitch). Excludes creators whose recent content matches the given keywords.

Added Discovery — audience targeting for Instagram creators (10k+ followers). Filter by audience.location, audience.interests, audience.brand_categories, audience.brands. Backed by four new classifier endpoints under /public/v1/discovery/classifier/: audience-locations, audience-interests, audience-brand-categories, audience-brand-names.

Added Discovery — exclude_handles parameter. Accepts up to 10,000 handles per request to exclude from search results. Values are normalized automatically.

Breaking Enrich by handle and batch enrichment — include_lookalikes default flipped from true to false. Same request body, different result set.

To restore previous behavior, set "include_lookalikes": true explicitly.

Deprecated Discovery — exclude_previous parameter. Use exclude_handles instead. Removed on Mar 12, 2026.

Added Post Details — content_type parameter now constrained to data | comments | transcript | audio. Instagram and TikTok support all four; YouTube supports data, comments and transcript.