ChatGPT Plugin

Clipit AI Connect

Explore your media monitoring

Brand: Unknown

Category: Pending
Primary Subcategory: Pending

Integration details

Description

Connect your Clipit account to explore your media monitoring in ChatGPT. View authorized accounts, inspect monitoring settings and saved charts, retrieve media statistics, find a small selection of permitted message examples with source links, and open the Clipit dashboard with relevant filters. You can also check your alert and newsletter subscription status and get help using the dashboard. Requires an existing Clipit account and authorization for the selected accounts. The connector reads existing data; it does not edit monitoring settings, change subscriptions or send messages. Message examples are bounded and do not replace the full filtered dashboard selection.

Integration type: Plugin
Verification status: Not applicable
Platform: ChatGPT
Category: Pending
Primary Subcategory: Pending
Secondary Subcategories: None listed
Brand: Unknown
Access: Account required
First tracked: 2026-10-15
Tool count: 25
Geography: US

Agent DiscoverabilityChatGPT Plugin Discovery Score

ChatGPT organic discovery is live

Your score is coming

ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.

ChatGPT recommends Airtable when a user asks it to update a CRM. — ChatGPT surfaces Airtable because it matches the user's request.

Primary Subcategory

Competitive lineup

No Primary Subcategory assigned yetOnce Clipit AI Connect has a Primary Subcategory, the Plugins and Connectors it competes with will appear here with their scores.

Tools

25 tools agents can invoke

Alerts ophalen
list_alerts
Alle accountgegevens en beschikbare filterkeuzes ophalen
get_account_masterdata
Berichten zoeken met filters of een letterlijke zoekterm; recente voorbeelden of hoogste PR-waarde/bereik. Combineer zoekacties naar behoefte; gebruik de gerapporteerde dekking om conclusies te begrenzen. Gebruik get_dashboard_link met dezelfde periode en filters om de volledige selectie te bekijken. RTV and podcast messages return one complete existing matched snippet, including any time codes; the 150-character cap applies only to other media. Do not reduce a returned media snippet to its time codes or claim that no text was supplied when it contains spoken text. No full transcript is fetched or assembled. A snippet can support a content claim only for what it actually says; a shared search match alone does not prove two items discuss the same story. For message detail questions use the returned source, author, date, sentiment, searchQuery/searchProfile and impactValue; never infer these from URLs. Label date simply Datum, never matchdatum. It is usually the publication date but can differ by source; explain that only when relevant, not as a routine caveat. Do not describe the returned metadata as all information stored by Clipit; retain the snippet and URL restrictions. For message top-N over a long period make one search_messages request for the full range: the server handles date windows, threshold selection, merging and deduplication within one shared analysisLimit (at most 500), returning at most 10 messages overall. Do not split the period into independent calls or add a PR/audience floor unless the customer requested that floor: highest_pr/highest_audience already try the highest supported threshold internally. A non-complete selection cannot establish an overall top-N; explain the actual coverage.reasons. The output limit alone does not invalidate a proven ranking. Always include each returned message's URL. Use recent examples for evidence, highest_pr or highest_audience for ranking; never infer a message maximum from daily totals. Do not repeatedly retrieve batches to bypass the 500-candidate budget. Use returned coverage.reasons and outputTruncated separately: output truncation alone does not invalidate an established maximum. Reputation analyses are available only when account hasReputationAnalysis is true. Impact data is available only when active impact rules exist. The report catalog lists general analysis types, not account permissions; get_report always requires an authorized account and the API enforces its capabilities. Unavailable metrics and reports are omitted or refused by the API. Never synthesize scores, infer missing metrics from other values, or substitute zero scores. Null means unknown, not absence. For distribution_excluded, explain briefly that messages are shielded because of a restriction on the source. For rankings say these are the highest values you can show, not a complete ranking; omitted messages may score higher. Only attribute coverage.omittedMessages entirely to this restriction when no other omission reason (missing_url or unavailable_messages) is present. If asked why, explain that Clipit shields these sources' message content and the connector therefore omits those messages; this restriction is separate from the customer's personal access rights. Do not speculate about licensing or suggest that changing dates, filters or permissions removes the source restriction. For editorial-media selections, consult the Editorial media definition in get_dashboard_knowledge article message-filters.
search_messages
Beschikbare accounts ophalen
list_accounts
Beschikbare analyses en grafieken verkennen
get_report_catalog
Beschikbare filters verkennen
get_filter_options
Broncategorieën ophalen
list_source_categories
Bronnen zoeken met een zoekterm of wildcard, bijvoorbeeld *krant*; maximaal 20 resultaten
search_sources
Bronthema's ophalen
list_source_themes
Compacte accountcontext: dashboardmodules en rechten, inclusief of de Nieuwsbriefmodule aanstaat. Geen zoekopdrachten, labels of landen-/talenlijsten. Gebruik aparte keuzetools voor filters. get_account_context is deliberately compact: account, modules, rights and support contacts. For saved searches/profiles/labels use their dedicated list tools or get_account_masterdata when a complete filter inventory is actually needed. Use known ISO country/language codes directly in filters (NL, DE; nl, en); do not retrieve global country/language lists by default. For module availability inspect dashboard.features; masterdata and source categories cannot establish it. When referring a customer to Clipit for a search request or support, use the assigned contacts from get_account_context: mediaMonitoringAdvisor is Media monitoring adviseur, mediaAnalyst is Media analist. Include the relevant assigned name and general phone/email; do not invent a missing name. Mention the visiting address only when relevant. Giving contact details is not drafting or sending a message. Confirm the relevant action before giving account-specific add/edit/delete steps. Null or missing capability values are unknown; generic documentation and enum choices do not establish permissions.
get_account_context
Create an authenticated dashboard navigation link. When the question's context calls for a graph, use page messages and analysis to preserve the metric AND breakdown from get_report, with the same dates and filters. Otherwise omit analysis; do not select an arbitrary graph. Choose a dashboardLinkSupported analysis from get_report_catalog. Proactively link messages for all results or download/export requests (multi-year supported). Opens the filtered set, not exactly ten examples; never exports or creates a public share. Other pages cannot receive message filters or analysis. For requests to see all messages, download them or export to Excel, proactively call get_dashboard_link with page messages, the established account, full requested date range and named filters, and include that link with the short steps from message-export in the first answer. This directly answers the request; it is not an unsolicited service tip. Preserve customer-requested filters; remove a PR/audience threshold added only for a previous top-N investigation, not an explicit customer constraint. If "all" is ambiguous, clarify the scope. One filtered dashboard link can cover multiple years: do not split it into yearly links or apply message-analysis/statistics limits to navigation. Only suggest splitting an export when its selection exceeds 65,535 messages. Select all filtered results, not just loaded rows, then Exporteren and confirm the count. Do not fetch repeated message batches to build a download. Do not volunteer unknown workbook columns, generic limitations or not-exported disclaimers; if specifically asked about columns, verify the export reference and state only what is established. Never imply the download was executed. A filtered result-set link is not a link to exactly the ten ranked examples; that separate feature is deferred. Use get_dashboard_link for supported customer-specific navigation; do not build guessed URLs from route templates. Its rejection of unsupported filters means no equivalent link was generated. When the question's context calls for a particular graph, use page messages with the same account, dates, filters and analysis as get_report. Choose a dashboardLinkSupported analysis from get_report_catalog, preserving both metric and breakdown: e.g. PR-waarde or PR-waarde - Mediatype. A follow-up "show this in the dashboard" can inherit that context. Omit analysis for general navigation; do not force a graph without a reason. This selects a graph, not a threshold filter or message sorting. Opening a saved analysis page does not override its charts.
get_dashboard_link
Dashboard reference: search by question, or read up to three articleIds from the index/results. Without either, list all pages and parts. Follow relatedArticles for details; generic documentation does not establish account permissions. Use this tool for dashboard operation, definitions and service explanations; it is not required to interpret search_messages results. For account-specific subscriptions use list_subscriptions; for metrics and messages use their data tools.
get_dashboard_knowledge
Grafiekpagina’s ophalen
list_chart_pages
Impact score: berekeningsregels ophalen
list_impact_calculation_rules
Labelcategorieën ophalen
list_label_categories
Labels ophalen
list_labels
Landen ophalen
list_countries
Landen zoeken
search_countries
Opgeslagen grafieken ophalen
list_saved_charts
Rapport of grafiek ophalen; kies een analyse uit get_report_catalog For analytics, use get_report_catalog to discover analysis names and supported intervals, then call get_report with the exact analysis name, account, explicit dates and relevant filters. All report types share this single tool; the old individual report tools are no longer published. Saved chart pages and personal graph-mail subscriptions are separate: use their discovery tools and list_subscriptions respectively. Resolve saved customer scope, use statistics for period-wide claims and messages for bounded evidence. Return at most ten messages with supplied URLs; analyze at most 500 internally per logical question. Snippets are not full articles. Explain actual completeness and do not claim maxima without established coverage. Statistics may run from the account start date, at most five calendar years back through today; get_account_context returns dashboard.statisticsPeriod. Hourly statistics remain limited to seven days. Reputation is optional and currently unavailable: account hasReputationScore is 0. Do not show reputation scores or drivers, infer them from impact, or substitute zero scores. A reputation-statistics request receives a specific API refusal. Impact is separate and remains available per message and in statistics. When impact.hasRules is false and discussing impact, state briefly that no impact rules are configured and the value is therefore very limited. Null means unknown, not absence. Use the selected account and filters consistently when comparing periods. Never infer causes solely from statistical differences. For editorial-media selections, consult get_dashboard_knowledge article message-filters.
get_report
Talen ophalen
list_languages
Talen zoeken
search_languages
Welke grafiekenmails, grafieken mailings, grafiekmailings, media updates of Analyse updates ontvang ik? Gebruik deze tool voor de exacte namen en persoonlijke abonnementsstatus. Ook redactionele nieuwsbrieven van Clipit, nieuwsselecties en alerts. Standaard alleen mijn abonnementen. Geen conclusie over bezorging of modulelicenties. For personal editorial newsletters, graph emails/media updates, news selections or alerts, use list_subscriptions and report the exact returned dashboard names and subscribed state. Module availability, editing rights and subscription status are different facts. The self-service Newsletter module does not establish editorial subscriptions; do not stop at module context when the customer asks what they receive. Grafiekenmailings, grafieken mailings, grafiekmailings, grafiekenmails and media updates are aliases for Analyse updates in subscription questions. "Welke heb ik dan?" continues the previously established subscription type and account scope. Call list_subscriptions; do not answer that the MCP cannot list these based only on knowledge or get_account_context. If list_subscriptions is missing from the client-visible tool list, state that the client tools need refreshing: the server supports it. A serverVersion response alone does not prove that the client has refreshed its tool descriptors. Never fabricate a call or a result.
list_subscriptions
Zoekopdrachten ophalen
list_search_queries
Zoekprofielen ophalen
list_search_profiles

FAQs