Redtrack MCP server
Yes. Redtrack publishes an official MCP server at https://mcp.redtrack.io/mcp. It's hosted. It offers 53 tools. We last checked it on 22 Nov 2026.
Official: answers MCP requests at mcp.redtrack.io, on Redtrack's own domain redtrack.io. Checked 22 Nov 2026. Maintained by Stuart Brameld.
Key facts
As of 22 Nov 2026, Redtrack's official MCP server scores 58/100 (provisional) on MartechDB's Agent Readiness Index (v1.1). It answered 55 of our 55 checks in the last 90 days and lists 53 tools.
How the Agent Readiness Index works · Wrong? Suggest a correction
Redtrack MCP server review
58/100
Redtrack's MCP server scores 58/100 on MartechDB's Agent Readiness Index: usable by AI agents, with gaps.
The good
Hosted by Redtrack: connect by URL, with nothing to install
Answered 100% of our checks in the last 90 days
The bad
Agent access costs more than the base plan
Not checked yet: how you sign in; a docs page listing its tools. Anything not checked scores nothing.
Official MCP server
Redtrack MCP server
- Address
- https://mcp.redtrack.io/mcp
- Runs
- Hosted
- Sign-in
- Cost to use
- Plan: Solo and above · Costs more than the base plan20 requests/minute; Premium MCP add-on ($1,990/year) gives 1,000 requests/minuteFrom Redtrack's own page
- Official MCP Registry
- Not listed
- How we know it's official
-
- Answers MCP requests at mcp.redtrack.io, on Redtrack's own domain redtrack.io.
- Tools
- 53 tools · listed 22 Nov 2026
- First seen
- 1 Oct 2026
- Last checked
-
Reachable ·
16 Nov 2026: reachable17 Nov 2026: reachable18 Nov 2026: reachable19 Nov 2026: reachable20 Nov 2026: reachable21 Nov 2026: reachable22 Nov 2026: reachable
- Uptime
- 100% · 55 of 55 checks reachable since 1 Oct 2026
Redtrack MCP server tools (53)
As the server lists them. Names and descriptions are the vendor's own.
-
getAnalyticsReportGet a traffic and conversion report: one row per combination of the `group` fields, with their values and the statistics. Pass `fields` to keep only the statistics you need; the group columns are always returned. Traffic report: one row per combination of the group fields, with the group field values and the statistics. The period is limited by the data retention of the plan. Returns: [Row], or {"items": [Row], "total": Row} when total=true. One report row, or the totals: the values of the group fields and the statistics. Every field is optional: only the requested group fields (and the IDs added for them) are present, the statistics are limited to those the shared-access user may see, fields removes the rest, and custom columns of the table settings are added under their IDs. Money is in the account currency; rates are fractions, not percent. Row fields: date (string): Group date - day, YYYY-MM-DD week (string): Group week - 'YYYY-MM-DD - YYYY-MM-DD', Monday to Sunday, in the account…
-
getCampaignsGet campaigns with statistics, one page at a time. The tool asks for page 1 of 100 campaigns unless told otherwise; raise `page` while `total_count` says more campaigns match. Without `status` deleted campaigns are included; pass status="1,2" for active and paused only. Returns: {"items": [Campaign], "total_count": integer or null}, plus "total" (the statistics summed over the whole selection) when total=true. Campaign fields: cache_buster_enabled (boolean) click_postbacks (array of object; may be null) cost_model (string; one of '', '-', 'CPA', 'CPL', 'CPC', 'CPM', 'RevShare', 'PopCPM', 'DoNotTrack', 'Hybrid'): Cost model; empty or "-" when not set. Not validated: other values are stored as sent coupon (string) cpc (number) created_at (string; read-only) creatives (array of object; may be null) custom_conv_type_conv_subs_geo_payouts (array of object; may be null) custom_conv_type_conv_subs_payouts (array of object; may be null) custom_conv_type_geo_payouts (array of object; may be nu…
-
getCampaignsShortList campaigns as short records — far smaller than getCampaigns, which returns whole campaigns with their streams and statistics. Use it to collect many campaigns at once, such as every campaign on a tracking domain before bulkUpdateEntities. The tool asks for 200 campaigns a page unless told otherwise, a size that fits in a tool result; pass the returned `next` back as `next` for the following page, it is null on the last one. Returns the campaigns (short DTO), ordered by id; of every status unless status is given (1 active, 2 paused, 3 deleted; another number is a 400, a non-numeric entry is ignored). Page size is 100 by default; outside 1-1000 the request fails with 400. In a shared-access session only the campaigns the shared user may see, limited both by campaign and by source. Returns: {"items": [CampaignShort], "total": integer — campaigns matching the filters on all pages, "next": string or null}. CampaignShort fields: id (string) title (string): The name of the campaign sourc…
-
getCampaignByIdGet a campaign by ID, with its streams, postbacks and payout settings. Statistics (stat) are filled only by getCampaigns. Returns: Campaign object: cache_buster_enabled (boolean) click_postbacks (array of object; may be null) cost_model (string; one of '', '-', 'CPA', 'CPL', 'CPC', 'CPM', 'RevShare', 'PopCPM', 'DoNotTrack', 'Hybrid'): Cost model; empty or "-" when not set. Not validated: other values are stored as sent coupon (string) cpc (number) created_at (string; read-only) creatives (array of object; may be null) custom_conv_type_conv_subs_geo_payouts (array of object; may be null) custom_conv_type_conv_subs_payouts (array of object; may be null) custom_conv_type_geo_payouts (array of object; may be null) custom_conversion_subs_payouts (array of object; may be null) custom_conversion_type_payouts (array of object; may be null) custom_geo_os_payouts (array of object; may be null) custom_geo_payouts (array of object; may be null) custom_pub_payouts (array of object; may be null) cu…
-
createCampaignCreate a campaign that sends the traffic of a traffic source through a tracking domain to streams. Creates a campaign for the traffic source source_id; the status defaults to active (1), and integrations.cost_update is turned off when the source has no cost-update integration. streams reference existing streams by id; a stream already used by another campaign is copied under a new id. The response is the stored campaign with its id and trackback_url. Create the streams first with createStream and send them as [{"id": "<stream id>", "weight": <integer>}]; a stream object in streams[].stream does not create a stream. Returns: The created campaign, with the fields getCampaignById returns; `id` identifies it and trackback_url is the tracking link to put in the traffic source.
-
updateCampaignUpdate a campaign. Updates a campaign. Only the fields sent in the body change; the rest keep their stored values. null clears an array but not a scalar field. A sent array replaces the stored one, but its items are decoded over the stored items with the same index, so fields missing in an item keep the values of that stored item. So a list field such as streams, postbacks or tags must be sent complete, not only the items to add or change: read it with getCampaignById first. Campaign streams reference existing streams by id: create the stream first with createStream, then send streams as [{"id": "<stream id>", "weight": <integer>}]. A stream object sent in streams[].stream does not create a stream; a stream already used by another campaign is copied, and the copy gets a new id. Returns: The updated campaign, with the fields getCampaignById returns.
-
updateCampaignsStatusActivate, pause or delete campaigns (status 3 deletes them). Sets one status on all the listed campaigns. All or none change - every id must be a campaign of the account (a repeated id counts as missing) and none of them may be banned. Returns: {"success": true}
-
getSourcesGet traffic sources with statistics, one page at a time. Active traffic sources with statistics; deleted (inactive) sources are not listed. The tool asks for page 1 of 100 sources unless told otherwise; raise `page` while `total_count` says more sources match. Returns: {"items": [Source], "total_count": integer or null}, plus "total" (the statistics summed over the whole selection) when total=true. Source fields: alias (string) campaign_count (integer; read-only) cost_id (string) cost_level (string; one of '', 'campaign', 'adset', 'ad', 'placement'): Level costs are updated at; availability depends on the source and the plan cost_models (array of string; may be null) created_at (string; read-only) currency (string) custom_rates (object) enable_direct_traffic (boolean) enable_parallel_tracking (boolean) external_id (string) external_id_alias (string) formats (array of string; may be null) google_analytics_key (string) id (string): Must match the path id in PUT imp_cost_id (string) imp_…
-
getSourceByIdGet a traffic source by ID, including a deleted (inactive) one. Statistics (stat) and campaign_count are filled only by getSources. Returns: Source object: alias (string) campaign_count (integer; read-only) cost_id (string) cost_level (string; one of '', 'campaign', 'adset', 'ad', 'placement'): Level costs are updated at; availability depends on the source and the plan cost_models (array of string; may be null) created_at (string; read-only) currency (string) custom_rates (object) enable_direct_traffic (boolean) enable_parallel_tracking (boolean) external_id (string) external_id_alias (string) formats (array of string; may be null) google_analytics_key (string) id (string): Must match the path id in PUT imp_cost_id (string) imp_id (string) integration_id (string) integration_types (object) integrations (object) pixels (array of object; may be null; read-only) pixels_relations (array of object; may be null): CAPI pixels attached to the traffic source. Google Analytics (GA4) pixels atta…
-
createSourceCreates a traffic source; the status defaults to active. On an account with a traffic-source limit a source above the limit is refused with 403. The response is the stored source with its id. Returns: The created source, with the fields getSourceById returns; `id` identifies it.
-
updateSourceUpdate a traffic source. The API replaces the whole source, so the tool reads the current source, applies `updates` and sends the result. Objects in `updates` are merged field by field; a list such as subs or formats is replaced as a whole: send the complete list. Setting status to active (or empty) restores a deleted source. Returns: The updated source, with the fields getSourceById returns.
-
deleteSourceDelete a traffic source. Soft delete: sets status to inactive. The source is still returned by id and can be cloned, and a PUT with an empty or active status restores it. Deleting it again returns 404. Returns: {"success": true}
-
cloneSourceClone a traffic source. Copies the source with " COPY" appended to its title. The integration is copied too for a google source with integration_id; other sources keep the integration_id of the original. Returns: The new source, with the fields getSourceById returns; `id` identifies it.
-
getSourcePresetsGet predefined traffic source templates/presets. These presets contain pre-configured settings for popular advertising platforms and can be used when creating new traffic sources. No parameters required. Returns: Array of SourcePreset objects: { "id": "string - Unique preset identifier (MongoDB ObjectId)", "title": "string - Display name of the traffic source preset (e.g., 'Meta (ex Facebook)', 'Google Ads', 'TikTok')", "type": "string - Source type: 'r' (recommended), 'i' (integrated with the platform API), 't' (template)", "alias": "string - Platform alias identifier (e.g., 'facebook', 'google', 'tiktok', 'taboola')", "url": "string - Official URL of the traffic source platform", "postback_url": "string - Pre-configured postback URL template for conversion tracking. Contains placeholders like {ref_id}, {offer_payout}, {status}", "postback_pixel": "string - Pixel tracking URL (if applicable)", "logo": "string - URL to the traffic source logo image", "imp_id": "string - Impression ID…
-
getOffersGet offers with statistics, one page at a time. The tool asks for page 1 of 100 offers unless told otherwise; raise `page` while `total_count` says more offers match. Statistics (stat) need both date_from and date_to, or time_interval=lasthour; otherwise stat is {}. Returns: {"items": [Offer], "total_count": integer or null}, plus "total" (the statistics summed over the whole selection) and "general_total" when total=true. Offer fields: cap (integer): Conversion cap, 0 - off. Conversions of the offer are counted over the calendar period of click_cap_period (day, week from Monday or month in the account time zone, any other value - the calendar year); once the cap is reached the offer is skipped in stream routing and its new conversions are rejected. cap_alert (boolean): Send an email to the account when the conversion cap is reached category (string; read-only): Set by the offer categorization, not by these operations; reset when the URL changes clcap (integer): Click cap, 0 - off. Cl…
-
getOfferByIdGet an offer by ID. Statistics (stat) and network_title are filled only by getOffers. Returns: Offer object: cap (integer): Conversion cap, 0 - off. Conversions of the offer are counted over the calendar period of click_cap_period (day, week from Monday or month in the account time zone, any other value - the calendar year); once the cap is reached the offer is skipped in stream routing and its new conversions are rejected. cap_alert (boolean): Send an email to the account when the conversion cap is reached category (string; read-only): Set by the offer categorization, not by these operations; reset when the URL changes clcap (integer): Click cap, 0 - off. Clicks sent to the offer are counted over the same calendar period as cap; once the cap is reached the offer is skipped in stream routing. clcap_alert (boolean): Send an email to the account when the click cap is reached click_cap (integer): Frequency cap per visitor, 0 - off: how many times one visitor (a unique user or an IP addre…
-
createOfferCreate an offer in an affiliate network (program_id). Creates an offer in the status passed in the body: 1 - active, 2 - paused, 3 - deleted. Without status (or with 0 or null) the offer is created active; any other status is rejected with 400. Returns: The created offer, with the fields getOfferById returns; `id` identifies it.
-
updateOfferUpdate an offer: pass only the fields to change. The API replaces the whole offer, so the tool reads the current offer, applies the changes and sends it back. Nested objects (payment, fingerprint_settings) are merged field by field; a list such as tags or country_codes is replaced as a whole, so send the complete list. Returns: The updated offer, with the fields getOfferById returns.
-
updateOffersStatusSet the status of several offers at once: activate, pause or delete them. Sets one status on all the listed offers (1 active, 2 paused, 3 deleted). Every id must be an offer of the account, otherwise nothing changes. Returns: {"success": true}
-
getLandingsGet landings and prelandings with statistics, one page at a time. The tool asks for page 1 of 100 landings unless told otherwise; raise `page` while `total_count` says more landings match. Returns: {"items": [Landing], "total_count": integer or null}, plus "total" (the statistics summed over the whole selection) when total=true. Landing fields: created_at (string; read-only) domain_id (string): Tracking domain of the landing; empty when none is set id (string; read-only) listicle (boolean) serial_number (integer; read-only) stat (object; read-only) tags (array of string; may be null) title (string): Trimmed; at most 200 bytes type (string; one of '', 'l', 'p'): l - landing, p - prelanding; empty if not set on create. The server does not validate it and returns what was stored. updated_at (string; read-only) url (string): Trimmed user_id (string; read-only)
-
getLandingByIdGet a landing or prelanding by ID. Statistics (stat) are filled only by getLandings. Returns: Landing object: created_at (string; read-only) domain_id (string): Tracking domain of the landing; empty when none is set id (string; read-only) listicle (boolean) serial_number (integer; read-only) stat (object; read-only) tags (array of string; may be null) title (string): Trimmed; at most 200 bytes type (string; one of '', 'l', 'p'): l - landing, p - prelanding; empty if not set on create. The server does not validate it and returns what was stored. updated_at (string; read-only) url (string): Trimmed user_id (string; read-only)
-
createLandingCreates a landing (type l) or a prelanding (type p). A domain_id, when set, must be a tracking domain. The response is the stored landing with its new id and serial_number; its statistics are empty. Returns: The created landing, with the fields getLandingById returns; `id` identifies it.
-
updateLandingUpdate a landing or prelanding. Updates a landing. Only the fields sent in the body change; the rest keep their stored values. A list field such as tags is replaced as a whole: send the complete list, not only the items to add. Returns: The updated landing, with the fields getLandingById returns.
-
deleteLandingDelete a landing or prelanding. This cannot be undone. Deletes the landing permanently and removes it from the landings and prelandings of every stream that uses it. Returns: {"success": true}
-
getNetworksGet affiliate networks (offer sources) with statistics, one page at a time. Active affiliate networks with statistics; deleted (inactive) networks are not listed. Returns an array by default, or {total, items} when total=true. The tool asks for page 1 of 100 networks unless told otherwise; raise `page` while `total_count` says more networks match. Statistics (stat) need both date_from and date_to, or time_interval=lasthour; otherwise stat is {}. Returns: {"items": [AffiliateNetwork], "total_count": integer or null}, plus "total" (the statistics summed over the whole selection) when total=true. Network fields: alias (string) cdp_hashing_personal_data_enabled (boolean) click_expiration (integer): Days after the click during which the network's conversions for it are accepted; 0 - no limit clickid (string) created_at (string; read-only) currency (string) custom_rates (object) customer_fields (array of string; may be null): Customer data fields; cleared when CDP is not enabled for the acc…
-
getNetworkByIdGet an affiliate network (offer source) by ID, including a deleted (inactive) one. offer_count and stat are filled only by getNetworks. Returns: AffiliateNetwork object, or {"error": "network not found"}. Network fields: alias (string) cdp_hashing_personal_data_enabled (boolean) click_expiration (integer): Days after the click during which the network's conversions for it are accepted; 0 - no limit clickid (string) created_at (string; read-only) currency (string) custom_rates (object) customer_fields (array of string; may be null): Customer data fields; cleared when CDP is not enabled for the account customer_subs (array of object; may be null) enable_ip_whitelist (boolean) event_tracking (boolean) hints (object) id (string; read-only) ip_whitelist (string) notes (string) offer_count (integer; read-only) offer_url (string) pixels (array of object; may be null; read-only) pixels_relations (array of object; may be null) postback_mode (string; one of '', 'AllowRepeat', 'DuplicateStatus',…
-
createNetworkCreate an affiliate network (offer source). Creates an affiliate network. The server sets status to active, generates postback_token and sets the five default customer subs; on ecom accounts it also creates the default store offer from offer_url. Returns: The created network, with the fields getNetworkById returns; `id` identifies it.
-
updateNetworkUpdate an affiliate network (offer source): pass only the fields to change. The API replaces the whole network, so the tool reads the current network, applies the changes and sends it back. Nested objects (custom_rates, hints) are merged field by field; a list such as subs or customer_fields is replaced as a whole, so send the complete list. Setting status to active restores a deleted network. Returns: The updated network, with the fields getNetworkById returns, or {"error": "network not found"}.
-
deleteNetworkDelete an affiliate network (offer source). Soft delete: sets status to inactive. The network is still returned by id, and a PUT with status active restores it. Ids that do not exist or belong to another account also return 204. Returns: {"success": true}
-
getNetworkPresetsGet list of pre-configured affiliate network templates (presets). These presets contain pre-filled settings for popular affiliate networks and can be used to quickly create a new network with correct postback URLs, click ID parameters, and other configurations. No parameters required. Returns: Array of NetworkPreset objects: { "id": "string - Unique preset identifier (MongoDB ObjectId)", "title": "string - Network name (e.g., 'Mobidea', 'Clickbank', 'Shopify')", "alias": "string - Preset alias identifier for special integrations (e.g., 'shopify', 'clickbank', 'woocommerce')", "url": "string - Network website URL", "logo": "string - URL to the network's logo image", "categories": ["string - Network categories (e.g., 'Pin-submits', 'Dating', 'E-commerce', 'Nutra', 'Gambling', 'Financial', 'Mobile Apps')"], "clickid": "string - Click ID parameter template used by this network (e.g., '{{EXTERNAL_ID}}', '{clickid}', '#s2#', '{aff_sub}')", "offer_url": "string - Default offer URL template w…
-
getDomainsGet domains of one type, one page at a time. Domains of the account of one type. total counts the domains matching type and url over the whole selection, not only the page. The certificate and private key of a custom certificate are never returned. Without `type` only tracking domains (track) are returned; pass pub or advert for the publishers or advertisers portal domains. The tool asks for page 1 of 100 domains unless told otherwise; raise `page` while `total` says more domains match. Certificate and private key material is never returned. Returns: {"items": [Domain], "total": integer}. Domain fields: id (string) user_id (string) url (string) type (string; one of 'track', 'pub', 'advert') fallback_url (string): Lowercased by the server created_at (string) ssl (object): Status of the domain SSL certificate. The certificate and its private key are never returned. use_auto_generated_ssl (boolean) acme (object): In the list, a copy of ssl active and expires, or an empty object until the…
-
createDomainCreate a tracking (track), publishers portal (pub) or advertisers portal (advert) domain. Exactly one SSL option must be enabled - use_auto_generated_ssl, or ssl.active with ssl.crt and ssl.key of a certificate that has not expired. With use_auto_generated_ssl the free certificate is issued in the background after the domain is saved. The response never contains ssl.crt or ssl.key. url is the host name only, e.g. track.example.com. Returns: The created domain, with the fields getDomains lists; `id` identifies it.
-
updateDomainUpdate a domain of any type. The API replaces the whole domain, so the tool reads the current domain, applies `updates` and sends the result. For a domain with use_auto_generated_ssl nothing about SSL has to be passed. A domain with a custom certificate keeps it when `updates` carries no ssl.crt and ssl.key; pass both to replace it, or to switch to a custom certificate (with use_auto_generated_ssl false). Returns: The updated domain, with the fields getDomains lists.
-
deleteDomainDeletes the domain permanently, together with its route and certificate in the gateway; a domain of another account is not found. Returns: {"success": true}
-
regenerateDomainSSLRegenerate the free SSL certificate of a domain with use_auto_generated_ssl. Starts issuing a new free SSL certificate for the domain. Returns: {"success": true}
-
bulkUpdateEntitiesSet the tracking domain on many campaigns or landings in one server-side call. To move entities from one tracking domain to another: 1. Resolve both domains. "dedicated" means the account's dedicated RedTrack subdomain, whose id is "none": it has no domain record. Look any other domain up with getDomains and its url filter (tracking domains only) and take the entry whose url is exactly that domain. A host getDomains does not find is the dedicated subdomain only if getCampaigns with domain_id="none", status="1,2,3" and per=1 returns a campaign whose trackback_url is on that host and whose domain_url is empty; otherwise stop and tell the user the domain was not found. 2. List the entities on the old domain, never without the domain filter: campaigns with getCampaignsShort (domain_id, status="1,2,3", per=200, passing next back until it is null), landings with getLandings (domain_id, per=50, page by page). domain_id="none" lists the entities on none of the account's tracking domains. A ba…
-
getStreamsGet streams, one page at a time. Template streams of the account (template=true); other streams are not listed. Streams that exist only inside campaigns are not returned. The tool asks for page 1 of 100 streams unless told otherwise; raise `page` while `total_count` says more streams match. Returns: {"items": [Stream], "total_count": integer}. Stream fields: expires_at (string) filters (object) id (string; read-only) landings (array of object; may be null) offers (array of object; may be null) prelandings (array of object; may be null) template (boolean) title (string) user_id (string; read-only)
-
createStreamCreate a stream (a funnel branch with offers, landings, prelandings and filters). Creates a stream. It is listed by GET /streams only with template true. Filters are keyed by name (filters.country, offers[].filters.os, ...); the tool fills a missing `kind` from the key, since the API applies a filter by its kind. Funnel (stream) filter kind: 0 country, 1 region, 2 city, 3 isp, 4 browser, 5 browser_version, 6 os, 7 os_version, 8 device_brand, 9 device_model, 10 ip, 11 subs, 12 language, 13 referrer, 14 domain_referrer, 15 device_type, 16 connection_type, 17 proxy_type, 18 fraud, 19 unique_visitor, 20 time_of_day, 21 day_of_week. The filter is applied by kind, not by the key it is stored under. Not validated: other values are stored as sent. Returns: The created stream, with the fields getStreams lists; `id` identifies it.
-
updateStreamUpdate a stream, including a stream used in a campaign. Updates a stream. Only the fields sent in the body change; the rest keep their stored values. null clears an array but not a scalar field. A sent array replaces the stored one, but its items are decoded over the stored items with the same index, so fields missing in an item keep the values of that stored item. The response is 201. Filters are keyed by name (filters.country, offers[].filters.os, ...); the tool fills a missing `kind` from the key, since the API applies a filter by its kind. Funnel (stream) filter kind: 0 country, 1 region, 2 city, 3 isp, 4 browser, 5 browser_version, 6 os, 7 os_version, 8 device_brand, 9 device_model, 10 ip, 11 subs, 12 language, 13 referrer, 14 domain_referrer, 15 device_type, 16 connection_type, 17 proxy_type, 18 fraud, 19 unique_visitor, 20 time_of_day, 21 day_of_week. The filter is applied by kind, not by the key it is stored under. Not validated: other values are stored as sent. Returns: The u…
-
deleteStreamDelete a stream. This cannot be undone. Removes the stream from every campaign that uses it and deletes it. If that leaves a campaign without streams, the stream is still deleted but the response is 400 with errors listing those campaigns. Returns: {"success": true}, plus "warnings" (the API's errors, naming the campaigns left without streams) when the stream was deleted but was the last stream of a campaign.
-
getStreamOptimizationGet the optimization settings of a stream. The stream as it is set in the first campaign that uses it: weight, optimization settings and the stream itself. 400: No campaign of the account uses the stream, or it could not be loaded (e.g. a template stream not added to any campaign). Returns: CampaignStream object: id (string) optimization (object) stream (object) weight (integer) optimization fields: clicks_limit (integer) conversion_types (array of string; may be null) count (integer) is_enabled (boolean) limit (integer) metric (string): Metric the optimization ranks paths by: cr, roi or epc (lowercase) multiplicator (number) subs (array of string; may be null) threshold (number) winner_share_limit (integer)
-
resetStreamOptimizationReset stream optimization statistics. Resets the collected optimization statistics of the stream. Returns: {"success": true}
-
getStreamOptimizationPathsGet stream optimization paths. Up to 500 funnel paths collected by the stream optimization, with the titles of their prelandings, landings and offers. Storage errors are not reported: the response is then 200 with fewer or no paths or without titles. Returns: Array of StreamPath objects: prelanding_id (string): Prelanding ID; a path without one repeats the value of the previous path landing_id (string): Landing ID; a path without one repeats the value of the previous path offer_id (string): Offer ID; a path without one repeats the value of the previous path sub{N} (string): N = 1..20. weight (integer): Share of the path in the traffic created_at (string): Second precision, in the server timezone prelanding (string): Prelanding title landing (string): Landing title offer (string): Offer title
-
getConversionsGet the conversions log, one page at a time. Conversions matching the filters, newest first. The period is limited by the data retention of the plan. Give date_from and date_to, or time_interval=lasthour. The tool asks for page 1 of 100 conversions unless told otherwise; raise `page` while `total` says more conversions match. Returns: {"items": [ConversionView], "total": integer}. ConversionView fields: attribution_type (string) browser (string) campaign (string): Title of the entity; empty if it no longer exists campaign_id (string) city (string) clickid (string): Click the conversion is attributed to connection_type (string) conv_time (string): Conversion time in the account timezone cost (number) cost_default (number) cost_source (number) country (string) coupon (string) created_at (string) currency (string) deduplicate_token (string) deeplink (string) default_type (integer; one of 0, 1): 1 - the conversion is of the default type device (string) device_brand (string) device_fullnam…
-
getClicksGet the clicks log, one page at a time. Clicks matching the filters, newest first. The period is limited by the data retention of the plan. Give date_from and date_to, or time_interval=lasthour; a valid clickid replaces the period. The tool asks for page 1 of 100 clicks unless told otherwise; raise `page` while `total` says more clicks match. Returns: {"items": [Click], "total": integer}. Click fields: adcreative_id (string): Not filled in the log: always empty browser (string) browser_fullname (string): Not filled in the log: always empty browser_version (string): Not filled in the log: always empty campaign (string): Title of the entity; empty if it no longer exists campaign_id (string) city (string) click_id (string): Not filled in the log: always empty connection_type (string) cost (number) cost_default (number): Not filled in the log: always 0 cost_source (number): Not filled in the log: always 0 country (string) created_at (string) date (string) deeplink (string) device (string)…
-
getCostUpdateLogsGet cost update sync logs (`/costs_logs`) — the raw record of what RedTrack fetched from the ad platform for each external ID at each sync cycle. Use this whenever you need to know: - What `source_cost` RedTrack received from the ad platform for a specific external ID (rt_campaign_id / rt_adgroup_id / rt_ad_id). - The cost update `level` (campaign / adgroup / ad) actually applied to a campaign. - The `source_timezone` (ad-platform account timezone) and `user_timezone` (RedTrack user timezone) — needed to diagnose timezone-mismatch discrepancies. - The sync `period` (5 / 15 / 30 / 60 / 360 minutes) and `created_at` timestamps — needed to verify sync regularity. - The `currency` and `source_alias` (facebook / google / tiktok / etc.) of the sync. Query parameters: date_from: Lower bound for `created_at`. Required. Format: YYYY-MM-DD. date_to: Upper bound for `created_at`. Required. Format: YYYY-MM-DD. campaign_id: Filter by one or more RedTrack campaign IDs. Optional. Comma-separated str…
-
getPostbackLogsGet server-to-server (S2S) postback logs (`/postbacks`) — the record of every postback RedTrack fired to an external destination (affiliate network, advertiser, or pixel endpoint) and how that destination responded. Use this whenever you need to know: - Whether a postback for a given click/conversion was actually delivered, and the HTTP `status` the destination returned (200 means the destination accepted it; anything else, or an `error`, indicates a delivery failure). - The exact `url` (and `request_body` for POST postbacks) that RedTrack sent — needed to diagnose malformed postback templates or missing macros. - Which `postback_type` fired (conversion / pixel / click / impression / first_click_postback) for a given `ref_id` or `external_id`. - The resolved `campaign`, `offer`, `network`, and `source` a postback belongs to, for auditing S2S delivery per entity. Query parameters: date_from: Lower bound for `created_at`. Required. Format: YYYY-MM-DD. date_to: Upper bound for `created_a…
-
getCostUpdateTroubleshootingGuideReturns a step-by-step diagnostic guide for investigating cost-update issues in RedTrack. Call this tool whenever the user reports any of the following: - Ad cost / spend in RedTrack does not match the ad platform (Facebook / Meta, Google Ads, TikTok, Snapchat, etc.). - Cost / spend numbers are missing for some campaigns, adgroups, or ads. - Cost update / auto cost sync appears broken, stuck, or out of date. - Total spend in RedTrack is lower than in the ad platform. - Discrepancy between RedTrack and Meta / Google / TikTok ads spend. - "RedTrack shows X but Facebook shows Y", "cost is wrong", "cost is not updating", "no cost data". The guide tells you which MCP tools to call (getCampaignById, getSourceById, getAnalyticsReport, etc.), in which order, with which parameters, and how to interpret the results. It also explains the diagnostic steps that cannot be performed with this MCP (internal cost update logs, ad platform API calls, integration token health) and tells you what to ask t…
-
updateIntegrationCampaignStatusPauses or activates a campaign on the ad platform (Ads Manager), through the integration connected to the source. Returns: {"success": true} on success.
-
updateIntegrationAdGroupStatusPauses or activates an ad group (ad set) on the ad platform (Ads Manager), through the integration connected to the source. Returns: {"success": true} on success.
-
updateIntegrationAdStatusPauses or activates an ad on the ad platform (Ads Manager), through the integration connected to the source. Returns: {"success": true} on success.
-
updateIntegrationCampaignDailyBudgetSets the daily budget of a campaign on the ad platform (Ads Manager), through the integration connected to the source. Returns: {"success": true} on success.
-
updateIntegrationAdGroupDailyBudgetSets the daily budget of an ad group (ad set) on the ad platform (Ads Manager), through the integration connected to the source. Returns: {"success": true} on success.
Connect your AI to Redtrack
Add the official Redtrack MCP server to Claude, ChatGPT or your editor.
Connect Redtrack MCP to Claude
Claude.ai and Claude Desktop
Open Customize, then Connectors.
Click +, then Add custom connector.
Name it Redtrack, paste the address below and click Add.
Connect Redtrack MCP to ChatGPT
Open Settings, then Apps, then Advanced settings.
Turn on Developer mode, then click Create.
Name it Redtrack and paste the address below.
ChatGPT usually only connects to servers with self-serve sign-in or none.
Connect Redtrack MCP to Cursor
Opens Cursor if it's installed. If nothing happens, use the JSON below.
Connect Redtrack MCP to VS Code
Opens VS Code if it's installed. Otherwise add the JSON settings from the Cursor tab to your .vscode/mcp.json.
Redtrack MCP server questions
Does Redtrack have an official MCP server?
Yes. Redtrack publishes its MCP server itself. Its address is on Redtrack's own domain (mcp.redtrack.io). MartechDB only lists servers a vendor publishes, never community copies.
What is the Redtrack MCP server URL?
The address is https://mcp.redtrack.io/mcp.
How do I connect Redtrack MCP to Claude, ChatGPT or Cursor?
In Claude, add https://mcp.redtrack.io/mcp as a custom connector (Customize, Connectors, Add custom connector). In Cursor, use the Add to Cursor link on MartechDB's page, or paste the JSON settings into your mcp.json. In Claude Code, run: claude mcp add --transport http redtrack https://mcp.redtrack.io/mcp
What can the Redtrack MCP server do?
It offers 53 tools, including getAnalyticsReport, getCampaigns, getCampaignsShort, getCampaignById and createCampaign (as listed on 22 Nov 2026). 23 are marked read-only and 22 can delete or overwrite data.
Is the Redtrack MCP server in the official MCP Registry?
No. Redtrack hasn't published it to the official MCP Registry.
Is the Redtrack MCP server working?
Yes. It answered our most recent check. We last checked it on 22 Nov 2026.
What are the alternatives to Redtrack's MCP server?
In Ad management & optimisation, the most agent-friendly alternatives with an official MCP server are Voluum (97/100), Blend AI (95/100), Birch (92/100), Groas (92/100) and Affise (91/100). Redtrack scores 58/100 on the same index.
Alternatives to Redtrack's MCP server
Vendors in Ad management & optimisation with an official MCP server, the most agent-friendly first. Redtrack scores 58/100.
Compare all 33 Ad management & optimisation vendors →Change history
What we've seen change for Redtrack, newest first.
-
Redtrack: Redtrack MCP server: 2 tools added (now 53)
-
Redtrack: Redtrack MCP server: 1 tool added (now 51)
Use with AI
Connect your AI to MartechDB
Then ask it about Redtrack, or any of the 1,850 marketing tools with an official MCP server. Free, no sign-in.
Claude.ai and Claude Desktop
Open Customize, then Connectors.
Click +, then Add custom connector.
Name it MartechDB, paste the address below and click Add. There's no sign-in.
The free plan allows one custom connector. On Team and Enterprise plans, an owner adds the connector first.
Claude Code
Run this in your terminal:
ChatGPT
Open Settings, then Apps, then Advanced settings.
Turn on Developer mode, then click Create.
Name it MartechDB, paste the address below and choose No authentication.
Developer mode is on Plus and Pro plans, and on Business, Enterprise and Edu if an admin allows it.
API
The same data as JSON, with no sign-in. See the OpenAPI description for every endpoint and field.
CLI
Search the directory and print the command to connect a vendor's server, from your terminal. Needs Node 18 or later. View on npm
Show that your MCP server is verified
If you work at Redtrack, add this badge to your MCP docs or README. It's free, it's the same for every vendor, and it updates from our daily checks: it shows uptime while the server responds, and says so if it stops.
Something wrong or out of date? Suggest a correction or email us, and we'll check it against the evidence. Every correction we make is listed on Corrections.