Integration

Mangools integration

Mangools is a suite of SEO tools. Research keywords with KWFinder (related, competitor and suggested keywords, metrics for your own keyword list, keyword gap, SERP and difficulty, saved keyword lists), read Google result pages with SERPChecker, backlinks and link metrics with LinkMiner (with CSV exports), domain authority, competitors and top pages with SiteProfiler, run SERPWatcher rank trackings (keywords, tags, annotations, email reports, rank statistics) and AI Search Watcher brand monitors, and look up location ids. Paid lookups draw from daily or monthly quota pools that get_account_limits shows; an identical request repeated within 24 hours is served from cache and not counted (keyword metric imports always count), and each pool allows at most 3 requests per 30 seconds. Left out: the deprecated keyword trends route, the audience route (always empty), the Looker Studio token check, the duplicate POST form of competitor keywords, the CSV downloads and the read-only share tokens of trackings and monitors. Authentication is the API key of the Mangools account (shown on its API token page), sent in the x-access-token header.

Connect Mangools How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Mangools.Built from the 78 Mangools tools below, on the connection you add once.

What you can do with Mangools

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

List, 5 operations

create_keyword_listCreate keyword listPOSTAgents: ask firstCreates a KWFinder keyword list with a name and keyword ids (from a keyword lookup; the list may start empty). An account holds up to 10,000 lists. Returns the list with its _id.
delete_keyword_listDelete keyword listDELETEAgents: ask firstPermanently deletes a KWFinder keyword list; the keyword data itself stays available to other lookups. No answer body.
get_keyword_listGet keyword listGETAgents: ask firstReturns one KWFinder keyword list (list) and the full data of every keyword in it: search volume and monthly history, CPC, PPC competition, difficulty and location.
list_keyword_listsList keyword listsGETAgents: ask firstLists your saved KWFinder keyword lists: _id, name, keyword ids, creator and timestamps. Pass a list _id to get_keyword_list for the keyword data.
rename_keyword_listRename keyword listPATCHAgents: ask firstRenames a KWFinder keyword list. No answer body.

Request, 5 operations

list_kd_url_lookupsList kd url lookupsGETAgents: ask firstLists your recent KWFinder URL authority lookups made with get_kd_url_metrics: url, the authority score (lps) returned then, and when. Free.
list_kwfinder_lookupsList kwfinder lookupsGETAgents: ask firstLists your last 100 KWFinder lookups, oldest first. t gives the type: 0 related keywords (kw, language, location, rank), 1 competitor keywords (url, location), 2 keyword gap. Free.
list_linkminer_lookupsList linkminer lookupsGETAgents: ask firstLists your recent LinkMiner backlink lookups: url, source filter (0 all, 1 new, 2 lost), links_per_domain and when. Free.
list_serpchecker_lookupsList serpchecker lookupsGETAgents: ask firstLists your recent SERPChecker lookups: keyword, location, platform, rank and when. Free.
list_siteprofiler_lookupsList siteprofiler lookupsGETAgents: ask firstLists your 50 most recent SiteProfiler lookups, newest first, as the URLs or domains analyzed under data. Free.

Tag, 5 operations

create_tagCreate tagPOSTAgents: ask firstCreates a keyword tag (name and color) on a SERPWatcher tracking; attach it to tracked keywords with assign_tags. Returns the tag with its _id.
delete_tagDelete tagDELETEAgents: ask firstPermanently deletes a tag from a SERPWatcher tracking and from every keyword that has it. No answer body.
list_all_tagsList all tagsGETAgents: ask firstLists every SERPWatcher keyword tag of the account, across trackings, newest first: _id, name, color and tracking id. Free.
list_tracking_tagsList tracking tagsGETAgents: ask firstLists the keyword tags of one SERPWatcher tracking, newest first: _id, name and color. Free.
update_tagUpdate tagPUTAgents: ask firstChanges the name and color of a SERPWatcher tag. Both are required: send the current value of the one you keep (from list_tracking_tags). No answer body.

Monitor, 4 operations

create_ai_monitorCreate ai monitorPOSTAgents: ask firstCreates an AI Search Watcher monitor that regularly asks the chosen AI models the given prompts and measures how often the brand is mentioned and cited. Counts against the plan's monitor and prompt allowances. Returns the new _id.
delete_ai_monitorDelete ai monitorDELETEAgents: ask firstPermanently deletes an AI Search Watcher monitor with its prompts, reports and tags; it stops collecting answers at once and its data is no longer returned. Answers an empty object.
get_ai_monitorGet ai monitorGETAgents: ask firstReturns analytics of one AI Search Watcher monitor over a date range (default the last 30 days): brand score, visibility and position, a time series, brand, prompt and domain rankings, and availableDates (the check days).
list_ai_monitorsList ai monitorsGETAgents: ask firstLists your AI Search Watcher monitors: _id, brand, domain, location and language ids, monitoring state (active, or warned or paused when the account no longer covers it) and the days with results. Pass an _id to get_ai_monitor.

Prompt, 4 operations

add_ai_monitor_promptsAdd ai monitor promptsPOSTAgents: ask firstAdds prompts to an AI Search Watcher monitor; its AI models are asked them from the next scheduled run, and they count against the plan's prompt allowance. Returns added_count and the added prompts with their ids.
delete_ai_promptsDelete ai promptsDELETEAgents: ask firstPermanently deletes AI Search Watcher prompts by id, across monitors, in one call; the monitors' other prompts keep running. Returns deleted_count.
get_ai_promptGet ai promptGETAgents: ask firstReturns metrics of one AI Search Watcher prompt over a date range: score, visibility, position, citation count and rank, a score time series, the brands and domains the AI answers mention most, and whether a Google AI Overview shows.
list_ai_monitor_promptsList ai monitor promptsGETAgents: ask firstLists the prompts of an AI Search Watcher monitor, each with its id, text, tags, timestamps, metrics for the date range (default the last 30 days) and Google AI Overview information. Pass a prompt _id to get_ai_prompt for detail.

Report, 4 operations

create_reportCreate reportPOSTAgents: ask firstSets up a scheduled email report on a SERPWatcher tracking: Mangools then emails the listed recipients weekly, monthly, or when ranks change. Returns the report with its _id.
delete_reportDelete reportDELETEAgents: ask firstDeletes a scheduled email report from a SERPWatcher tracking; its recipients stop receiving it. No answer body.
list_reportsList reportsGETAgents: ask firstLists the scheduled email reports of a SERPWatcher tracking, newest first: _id, type (0 rank-change alert, 1 weekly, 2 monthly), name, triggers and recipient emails (personal data).
update_reportUpdate reportPUTAgents: ask firstChanges an email report of a SERPWatcher tracking. name, type and emails are required: read them with list_reports and send them back with the change. Mangools emails the recipients listed. No answer body.

Tracked keyword, 4 operations

add_tracked_keywordsAdd tracked keywordsPOSTAgents: ask firstAdds keywords to a SERPWatcher tracking (ones already tracked only get the given tags); new ones are queued for rank checks and count against the tracked-keyword allowance. tag_ids is required and may be empty.
get_tracked_keywordGet tracked keywordGETAgents: ask firstReturns the rank detail of one tracked keyword over a date range: ranking URLs (organic, Map Pack, Featured Snippet) with daily ranks, visual metrics and estimated visits, search volume, snapshot links and tags.
list_tracked_keywordsList tracked keywordsGETAgents: ask firstLists the tracked keywords of a SERPWatcher tracking: keyword id, keyword, location and platform, with search volume. fields (comma-separated) chooses the attributes and can add rank, active and history.
remove_tracked_keywordsRemove tracked keywordsDELETEAgents: ask firstStops tracking keywords: removes them, by tracked keyword id, from a SERPWatcher tracking, whose statistics are then recomputed. No answer body.

Tracking, 4 operations

create_trackingCreate trackingPOSTAgents: ask firstCreates a SERPWatcher rank tracking of a domain for keywords in one location and platform; the keywords are queued for rank checks and default weekly and monthly email reports are set up. Uses the tracking and keyword allowance.
delete_trackingDelete trackingDELETEAgents: ask firstPermanently deletes a SERPWatcher tracking with all its tracked keywords and email reports. No answer body.
list_trackingsList trackingsGETAgents: ask firstLists your SERPWatcher rank trackings, newest first: _id, domain, location, platform, keyword count and ids, local-SEO business and active reports. Set is_with_deleted to true to include deleted trackings.
update_trackingUpdate trackingPATCHAgents: ask firstChanges a tracking's domain or its local-SEO business (Google Business Profile place_id, name, address, ludocid). domain is always required: send the current one (from get_tracking) to keep it. Fails with 409 if another active tracking covers it.

Annotation, 3 operations

create_annotationCreate annotationPOSTAgents: ask firstAdds a dated note to the timeline of a SERPWatcher tracking, shown on its rank charts (a site launch, a campaign, an algorithm update). Returns the annotation with its _id.
delete_annotationDelete annotationDELETEAgents: ask firstPermanently deletes an annotation from the timeline of a SERPWatcher tracking. No answer body.
update_annotationUpdate annotationPUTAgents: ask firstChanges the text of an annotation on a SERPWatcher tracking timeline; date and nonShareable can change too. No answer body.

Favorite, 3 operations

delete_favorite_backlink_listDelete favorite backlink listDELETEAgents: ask firstPermanently deletes a favorite-backlink list; the backlinks themselves are unaffected. No answer body.
get_favorite_backlink_listGet favorite backlink listGETAgents: ask firstReturns one favorite-backlink list with the full data of each starred backlink: source and target URL, anchor, link type, first and last seen, Citation and Trust Flow.
list_favorite_backlink_listsList favorite backlink listsGETAgents: ask firstLists your favorite-backlink lists (one per target domain), most recently updated first, each with its domain, starred link ids and links. Free.

Keyword, 3 operations

add_keywords_to_listAdd keywords to listPOSTAgents: ask firstAdds keywords, by KWFinder keyword id, to a keyword list; a keyword already in it is not added twice, and a list holds at most 10,000 keywords (more answers HTTP 422). No answer body.
get_keywords_by_idGet keywords by idPOSTAgents: ask firstReturns the stored metrics (search volume, monthly history, CPC, PPC competition, difficulty, location, language) of keywords by their KWFinder _id from earlier lookups, up to 10,000 ids per call; unknown ids are dropped. Free.
remove_keywords_from_listRemove keywords from listDELETEAgents: ask firstRemoves keywords, by KWFinder keyword id, from a keyword list; ids not in the list are ignored. No answer body.

Url metric, 3 operations

get_kd_url_metricsGet kd url metricsGETAgents: ask firstReturns the URL authority score (lps) KWFinder uses as an input to keyword difficulty, for one URL. Billed from the daily kw-url-metrics pool (an identical request within 24 hours is served from cache, free).
get_linkminer_url_metricsGet linkminer url metricsGETAgents: ask firstReturns Majestic link metrics (Citation Flow, Trust Flow, referring domains, backlinks) of a URL, domain or subdomain, as a list. Billed from the daily lm-url-metrics pool (an identical request within 24 hours is free).
get_serpchecker_url_metricsGet serpchecker url metricsGETAgents: ask firstReturns authority metrics of one URL: Moz Domain and Page Authority, Majestic Citation and Trust Flow, Facebook engagement, the Mangools authority rank and TopRank. Billed from the daily sch-url-metrics pool (repeat within 24 h free).

Export, 2 operations

create_backlink_exportCreate backlink exportPOSTAgents: ask firstStarts a CSV export of the backlinks of a target; the url must match a target from suggest_backlink_export_targets. Charged to the Backlink rows allowance: the suggested cost, or limit if lower. Poll list_backlink_exports for the file.
list_backlink_exportsList backlink exportsGETAgents: ask firstLists your backlink CSV exports, newest first: target, filters, cost, state (600 done) and, once done, downloadURL, a link to the CSV given for 60 days after completion. Poll it after create_backlink_export. Free.

Location, 2 operations

get_locationGet locationGETAgents: ask firstReturns one location by id: name, canonical name, label, country code, target type, parent location id and Google domain. Fails with 404 for an unknown id. Free.
search_locationsSearch locationsGETAgents: ask firstFinds location ids by name (country, region or city), up to 100 active matches, each with _id, label, country code and target type. Pass _id as location_id to the keyword, SERP, tracking and AI monitor tools. Free.

Serp, 2 operations

get_keyword_serpGet keyword serpGETAgents: ask firstReturns a keyword's SEO difficulty and its Google results page: organic results with authority metrics, result count, click-through rate per position, search intent and SERP features. Billed: one SERP lookup (repeat within 24 h free).
get_serpGet serpGETAgents: ask firstReturns the Google results page of a keyword in a location with all features: organic and paid results with authority metrics, visibility and CTR, result count and serp_snapshot_id. Billed: one SERP lookup (repeat within 24 h free).

Settings, 2 operations

get_ai_monitor_settingsGet ai monitor settingsGETAgents: ask firstReturns the settings of an AI Search Watcher monitor: brand, brand variations, domain, location and language ids, AI model ids, active email reports (3 weekly, 4 monthly) and whether status notification emails are on.
update_ai_monitor_settingsUpdate ai monitor settingsPUTAgents: ask firstChanges an AI Search Watcher monitor's brand, brand variations, domain, AI models, email reports or dashboard and notification flags; only the fields sent change. Location, platform and language cannot be changed.

Assign, 1 operation

assign_tagsAssign tagsPOSTAgents: ask firstAttaches tags to tracked keywords of a SERPWatcher tracking, keeping the tags they already have. No answer body.

Backlink profile, 1 operation

get_backlink_profileGet backlink profileGETAgents: ask firstReturns the backlink profile of a URL or domain: Majestic totals, a calendar of new and lost links, top referring domains and anchor texts. Needs get_site_overview for the same url within 24 hours, else it fails with Not in history. Free.

Competitor, 1 operation

list_site_competitorsList site competitorsGETAgents: ask firstLists competitor domains of a URL or domain with their TopRank, Facebook engagement, referring IPs and a score. Needs get_site_overview for the same url within 24 hours, else it fails with Not in history. Free.

Competitor domain, 1 operation

list_competitor_domainsList competitor domainsGETAgents: ask firstLists domains that compete with a URL or domain for organic keywords, as domain names under data. Needs a list_competitor_keywords lookup for the same url and location_id in the last 24 hours, else it fails with No result. Free.

Competitor keyword, 1 operation

list_competitor_keywordsList competitor keywordsGETAgents: ask firstLists the keywords a URL or domain ranks for in a location, with search volume, monthly history, CPC, PPC competition and difficulty, ordered from its worst position to its best. Billed: one Keyword lookup (repeat within 24 hours free).

Detail, 1 operation

get_trackingGet trackingGETAgents: ask firstReturns one SERPWatcher tracking and its tracked keywords: keyword id, keyword, current ranking URL, tags and last check time, plus white-label settings. Rank history and statistics are in get_tracking_stats.

Favorite link, 1 operation

set_favorite_backlinkSet favorite backlinkPATCHAgents: ask firstStars or unstars a backlink (its _id from list_backlinks) and returns the favorite list of its target domain with the starred links. An unknown link id is ignored and the answer is empty.

Gap analysis, 1 operation

analyze_keyword_gapAnalyze keyword gapPOSTAgents: ask firstFinds keywords that up to 5 competitor domains rank for and your domain does not, one result per competitor, with search volume, difficulty, CPC and the competitor position. Billed: one Keyword lookup per competitor (repeat within 24 h free).

Generate, 1 operation

generate_ai_promptsGenerate ai promptsPOSTAgents: ask firstSuggests monitoring prompts for a brand and a short summary of how AI models describe it (perception), to pass to create_ai_monitor. Only brand and language_id shape the answer; location_id, platform_id and domain are required but unused.

Keyword import, 1 operation

get_keyword_metricsGet keyword metricsPOSTAgents: ask firstReturns metrics for the keywords you list, in one location and language: average and monthly search volume, CPC, PPC competition and cached difficulty, each with a keyword _id. Billed: one Keyword lookup per call, also when repeated.

Limit, 1 operation

get_account_limitsGet account limitsGETAgents: ask firstReturns the quota pools of the account with what is left: per pool limit, remaining and seconds until reset, plus plan level, tracking and prompt allowances and the per-request row caps. Free; call it before costly lookups.

Link, 1 operation

list_backlinksList backlinksGETAgents: ask firstLists backlinks of a URL, domain or path, up to 500 per page (page from 0): source and target URL, anchor, link type, first and last seen, Citation and Trust Flow. Billed: each link returned uses one Backlink row.

Model, 1 operation

list_ai_modelsList ai modelsGETAgents: ask firstLists the AI models AI Search Watcher can ask (ChatGPT, Gemini, Claude and others): numeric id, model, display name and active, deprecated, citation and SERP flags. Pass ids of models that are not deprecated to create_ai_monitor.

Multiple tracking, 1 operation

create_trackingsCreate trackingsPOSTAgents: ask firstCreates SERPWatcher rank trackings of one domain and keyword set for each location and platform pair not tracked yet; each gets all the keywords and default weekly and monthly email reports. Uses the tracking and keyword allowance.

Overview, 1 operation

get_site_overviewGet site overviewGETAgents: ask firstReturns authority metrics of a URL or domain: Moz Domain and Page Authority, Majestic Citation and Trust Flow, referring IPs, TopRank and Facebook shares, with history since 2016 if available. Billed: daily sp-overview pool (repeat in 24 h free).

Related keyword, 1 operation

list_related_keywordsList related keywordsGETAgents: ask firstLists keyword ideas related to a seed keyword (the seed first, the rest unsorted) with search volume, monthly history, CPC, PPC competition and difficulty. Billed: one Keyword lookup (an identical request within 24 hours is free).

Reset, 1 operation

refresh_serpRefresh serpGETAgents: ask firstRe-fetches serp_count Google result pages (1 to 10, default 1) of a keyword from page on, bypassing the cache. Billed on every call: one SERP lookup per page returned with results. Use get_serp for cached data.

Snapshot, 1 operation

get_serp_snapshotGet serp snapshotGETAgents: ask firstReturns a screenshot of a stored Google results page as a base64 PNG (image). Pass serp_snapshot_id from get_serp or get_keyword_serp exactly as returned, never the bare _id; a null serp_snapshot_id has no snapshot. Free.

Stats, 1 operation

get_tracking_statsGet tracking statsPOSTAgents: ask firstReturns rank statistics of a SERPWatcher tracking over a date range: per keyword current, best and average rank, rank change and history, estimated visits, Map Pack and Featured Snippet positions and SERP features, plus annotations.

Suggest, 1 operation

suggest_backlink_export_targetsSuggest backlink export targetsPOSTAgents: ask firstLists the export targets for a URL (root domain, subdomain, exact URL, protocol variants) with backlink and referring-domain counts and the estimated export cost in backlink rows. Billed from the daily lm-suggest pool.

Suggested keyword, 1 operation

list_suggested_keywordsList suggested keywordsGETAgents: ask firstLists keywords suggested from those a URL or domain ranks for (with position and search volume), meant for seeding a SERPWatcher tracking. Billed: one Keyword lookup (an identical request within 24 hours is free).

Unassign, 1 operation

unassign_tagsUnassign tagsPOSTAgents: ask firstDetaches tags from tracked keywords of a SERPWatcher tracking; tags not attached are ignored. No answer body.

Use Mangools from Claude Code, Cursor or Codex

Connect Mangools in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Mangools operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Mangools

Create your workspace
Sign up free. Nothing to install.
Connect Mangools once
Add your Mangools connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.