MCP App Store
by KeenEthics
Roivenue icon

Roivenue MCP Server for Claude and ChatGPT

By Roivenue s.r.o.

Overview

Roivenue is a marketing-attribution and data-integration platform. This connector lets you ask your own Roivenue data questions in plain language. Ask how your channels and campaigns performed (revenue, cost, ROMI, conversions) under any attribution model, including Roivenue's AI Data Driven model. Explore customer journeys: typical path length, days to convert, which channels appear in the same converting paths, and the touchpoint-by-touchpoint story behind a single order. Check the health of your data sources and see when your data was last refreshed. Answers use the same defaults as your Roivenue app — your attribution model, currency and date window — so the numbers match what you see there, and the connector can hand you a deep link that opens the exact same view in Roivenue. You sign in with your own Roivenue credentials (OAuth 2.1); the connector sees only the accounts and data your login already has access to, and every tool is read-only — the single exception records optional thumbs-up/down feedback. Customer-identifying fields are removed from journey and order data by design. Product-knowledge questions are answered from Roivenue's official documentation, with sources cited. Requires a Roivenue subscription and login (app.roivenue.com).

How to connect

  1. Claude: Open the Roivenue listing in the Claude connectors directory and choose Connect. Claude listing Roivenue exposes 21 tools here.
  2. ChatGPT: Open the Roivenue listing in the ChatGPT app directory and choose Connect. ChatGPT listing Roivenue exposes 27 tools here.

Tools

account_markets

ChatGPT
The account's markets (data partitions) with each one's reporting currency and data coverage. A multi-market account returns one row per currency from query_performance — use this to label each row in its market's currency (never compare across currencies) and to sanity-check date ranges against actual coverage. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

account_overview

ChatGPT
Start here once an account is chosen. One call that orients you to the account: its name, the queryable catalog (measures, dimensions with their side — cost|web|both — and custom/built_from, attribution model codes, scope dimensions, default metric) and connector health. Read dimension_guide: attributed measures are 0 on a cost-side dimension such as adCampaign. Use the codes it returns to plan list_dimension_values / query_performance. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

attribution_model_settings

ChatGPT
The settings behind the account's attribution models: for each model its code, algorithm (the AI Data-Driven model is rnnProbability), whether it is the user's default, the journey (path) windows — max days between touchpoints, max journey length, path stitching, synthetic impressions — the AI model's training parameters (model period, training data period, scoring), time-decay half-life, touchpoint weight rules and UMM mixing ratio; plus each site's attribution windows (path cut-off days). Every value says whether it was set explicitly or is the pipeline default. windows_guide says which window is which. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

connector_health

ChatGPT
Report an account's data-source (connector) health: per-connector-type source counts (active/inactive), each type's data-coverage window, failing sources with their last error message, and an auth-expiry hint. Use this to explain data gaps — missing spend, untracked conversions, a channel showing zero — from connector status instead of guessing. For per-job refresh timestamps and error drill-down use data_freshness. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

conversion_detail

ChatGPT
The per-touchpoint journey behind ONE conversion/order: pass conversion_id = a pathConversionId from transaction_history verbatim — often a string like 'OBJ26113784'; never strip the prefix. Returns touchpoints in time order (keys are dataset column codes; customer-/email-identifying columns are removed); a no-match id returns empty touchpoints plus a note. Takes NO dates or filters by design — it looks the conversion up across all time. attribution_model takes a model code, never a display name. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

create_custom_dimension

ChatGPT
CREATE a custom (derived) dimension with rules in the user's Roivenue account — ONLY on an explicit ask, after preview_custom_dimension was shown to the user. Needs the Operator role (clear 403 otherwise). Triggers a data recalculation up to a year back — always relay recompute_note and explorer_url. Refuses while the account's ETL jobs are running (wait, watch data_freshness, retry) and after 5 creates on the account within an hour. Args as in preview_custom_dimension (minus dates). Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

create_dashboard

ChatGPT
CREATE a custom dashboard in the user's Explorer app — ONLY on an explicit ask (it writes to their account; private to them until shared; the app may notify them). Widgets (max 12): 'kpi-card' {metric}; 'dimension-main-chart'|'dimension-pie-chart'|'dimension-horizontal-bar-chart' {metric, dimension}; 'dimension-table' {metrics (1-2), dimension}; 'table-report' {table: {rows, values, columns?, view?, date_axis?} as in create_table_report — a snapshot tile, no report is saved}; other widget types must be added in the app. Metric/dimension/attribution take codes from describe_metrics — unknown values are REJECTED (the API would store a silently blank tile); near-misses auto-correct, echoed in resolved. Windows are RELATIVE so the dashboard stays current: from_type ('last'|'this'|'previous'), period ('day'|'week'|'month'|'quarter'|'year'), multiplicand, granularity ('daily'|'weekly'|'monthly'); omit window/attribution_model/granularity for the caller's Explorer defaults (echoed in defaults_applied). Filters as in query_performance; layout is automatic. Returns {id, url, widgets} — give the user the url as a Markdown link. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

data_freshness

ChatGPT
Per-pipeline data freshness for the account: one row per ETL job with status (notRun/waiting/processing/completed/failed/canceled), last_completed_at (the 'data last refreshed' timestamp), the unfinished-data window (missing since), and the last error message. Use for 'when was my data last refreshed?' / 'why is a source stale?'. Narrow with connector_type or only_problems=true; limit default 100 (max 200, _truncated sentinel when more exist). Pass etl_job_id (from a row) to drill into that job's last error — returns the exception message plus an error_mitigation_message fix hint (quote it to the user); raw stack traces are never returned. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

describe_metrics

ChatGPT
The queryable catalog: measures, dimensions, attribution models, and default metric for the account (account_overview wraps this for a one-call orientation). Each attribution model includes its code — pass that code (not the display name) to query_performance. Each dimension carries side (cost = platform-only, attributed measures always 0; web; both) and, for account-built ones, custom + built_from — see dimension_guide. scope_dimensions flags dimensions (e.g. site) that materially scope a result; use list_dimension_values to see their values. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

get_skill

ChatGPT
Load focused guidance for the kind of task you are about to do, and follow it. Call this BEFORE answering when the question matches one of the skills — path-analysis (customer-journey / touchpoint-path questions, channel overlap, order drill-in), reconciling-numbers (the user's figure disagrees with yours), attribution-and-dimensions (zero revenue with non-zero cost, or a name that could live under several dimensions), budget-reallocation (where to shift spend), data-health (stale or broken-looking data), charts (an explicit chart request). Needs no account or token. Most questions need no skill at all.

list_accounts

ChatGPT
List the Roivenue accounts this connection can access, as [{id, name}]. Call this when no account is pinned by server config: present the options and ask the user which account to explore, then pass the chosen id as account_id to the other analyst tools. If an account is pinned server-side you don't need this — the tools use it directly. In answers and citations name accounts by NAME — ids are internal.

list_dimension_values

ChatGPT
List the distinct values of a dimension for the account (e.g. the sites in site), ranked by measure (default netRevenue) over [date_from, date_to]. Use this to enumerate scope options (sites, markets, conversion types) before querying or before asking the user to choose. Omitted attribution_model defaults to the user's Explorer model; without one, attributed measures sum across EVERY model and read ~Nx high. Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

my_settings

ChatGPT
Call this FIRST, before your first data answer, to mirror the calling user's Explorer defaults: their selected attribution model, currency, default date window (as an object plus human-readable text) and time granularity, and the accounts they can reach. Compute data answers with these defaults so the numbers match what the user sees in the Explorer app — and say so when you do. Works with no account selected (it is about the user, not an account); account_id is optional and used only to resolve the attribution model to its canonical code. Name accounts by NAME in answers, never ids.

path_histogram

ChatGPT
A distribution histogram from the Explorer Path analysis page for an account over [date_from, date_to]. kind: 'days_to_convert' (first touch to conversion) or 'touchpoints_to_convert' (touchpoints per converting path). Returns {kind, bins:[{title, amount, share}]}. Bins are FIXED server-side — days: 1, 2, 3-7, 8-15, 16-30, 31+; touchpoints: 1, 2, 3, 4, 5-10, 10+ — custom bins are not possible. attribution_model takes a model code from describe_metrics, never a display name (the AI Data Driven code varies by account: 'RnnProbability' or 'aiDataDriven') — and must be MULTI-TOUCH: under last-touch every path collapses to one touchpoint and path metrics are degenerate. An empty result carries a note — follow it. Money is EUR-pinned. Takes no toggles. Scope with filters like [{"name": "platformName", "values": ["Google Ads"]}] (optional "operator": in|notIn|contains|notContains; contains takes one "value"). Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

path_journeys

ChatGPT
The Explorer Path analysis page's path table — which converting journeys drive conversions for an account over [date_from, date_to], grouped by a dimension. group_by is a conversionPaths dimension code (e.g. platformName) labelling each touchpoint. measures default [conversions, conversionPathLength, netRevenue]: pass PRIMARY measure codes only (conversions, netRevenue, grossProfit, visits, impressions) plus the pseudo-measure conversionPathLength (not in the catalog); derived ratio measures (conversionRate, averageOrderValue, romi) ERROR here; max 4. Returns {rows, total_row_count, aggregates}, each row a path in touch order. limit default 20, max 50, page with offset; sort via sort_by (measure code or conversionPathLength) + sort_dir; journey_contains keeps only paths containing that value. attribution_model takes a model code from describe_metrics, never a display name (the AI Data Driven code varies by account: 'RnnProbability' or 'aiDataDriven') — and must be MULTI-TOUCH: under last-touch every path collapses to one touchpoint and path metrics are degenerate. An empty result carries a note — follow it. currency defaults EUR. Scope with filters like [{"name": "platformName", "values": ["Google Ads"]}] (optional "operator": in|notIn|contains|notContains; contains takes one "value"). Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

path_kpis

ChatGPT
The Explorer Path analysis page's KPIs for an account over [date_from, date_to]: average path length, average days to convert, the share of conversions the attribution model redistributes away from last-touch, and the touchpoint-type mix (amount plus a 0-1 share). attribution_model takes a model code from describe_metrics, never a display name (the AI Data Driven code varies by account: 'RnnProbability' or 'aiDataDriven') — and must be MULTI-TOUCH: under last-touch every path collapses to one touchpoint and path metrics are degenerate. An empty result carries a note — follow it. Money is EUR-pinned. Visit is the effective click+visit bucket (the ETL remaps clicks to Visit, so Click is ~0). exclude_stitched_paths defaults true to match the app's fresh-open view; set it false to include cross-device-stitched journeys (which raises the redistribution share). include_impressions / include_synthetic_impressions add those buckets and raise the totals. Scope with filters like [{"name": "platformName", "values": ["Google Ads"]}] (optional "operator": in|notIn|contains|notContains; contains takes one "value"). Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

path_overlap

ChatGPT
The Explorer Path analysis overlap (Venn) widget: how often a dimension's values co-occur in the same converting paths over [date_from, date_to]. Give a dimension (a conversionPaths dimension code such as platformName) and 2-3 values — the endpoint is exponential in the value count, so it caps at 3. Take candidate values from list_dimension_values, but they must match EXACTLY as they appear in paths (that tool reads the performance dataset by default). Returns {dimension, values, subsets:[{sets, size}] one per non-empty subset, insights:[...]} where an insight reads like "12.3% of 'Google Ads' paths also contain 'Facebook Business'". attribution_model takes a model code from describe_metrics, never a display name (the AI Data Driven code varies by account: 'RnnProbability' or 'aiDataDriven') — and must be MULTI-TOUCH: under last-touch every path collapses to one touchpoint and path metrics are degenerate. An empty result carries a note — follow it. Money is EUR-pinned. Scope with filters like [{"name": "platformName", "values": ["Google Ads"]}] (optional "operator": in|notIn|contains|notContains; contains takes one "value"). Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

preview_custom_dimension

ChatGPT
Validate and DRY-RUN a draft custom (derived) dimension on live data — nothing is written. Invalid drafts are REJECTED with the valid options; returns per-value coverage vs the 'Undefined' bucket, rule conflicts, and sample rows. when = OR-list of AND-groups; operators: contains|startsWith|endsWith|equals|regex|in|notIn|isEmpty|isNotEmpty|not-variants|greaterThan|lowerThan|greaterOrEqualTo|lowerOrEqualTo (case-INSENSITIVE; only equals honors case). then.type: constant|other_dimension|expression|regex_group. Conflicting rows: preview shows the LAST matching rule, production keeps the FIRST. Run this and show coverage/conflicts before create_custom_dimension. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

query_performance

ChatGPT
Query an account's marketing performance: aggregate measures by group_by dimensions over [date_from, date_to] (ISO-8601). Use describe_metrics for valid codes; pass attribution model codes (e.g. 'LastTouch') in attribution_models. Scope with filters shaped like [{"name": "site", "values": ["<a site value>"]}]; operator contains/notContains takes one value (a filter with the wrong key is rejected, never silently ignored). Leave granularity unset for a single total per group — it defaults to 'aggregate'; the API's own default is per-DAY rows, so only set granularity to 'Daily'/'Weekly'/'Monthly' when you actually want a time series; 'Weekly'/'Monthly' cover whole weeks/months, so the window is widened to those boundaries for you and the widening reported in a _window_snapped row — report THAT window, and don't retry with hand-trimmed dates. A Daily/Weekly/Monthly series returns up to 400 rows in one call — never chunk a year of days into multiple queries. A _suspect row (kind cost_side_dimension, with alternatives, or zero_attribution) means the result is the wrong-dimension/wrong-attribution signature, NOT real numbers: re-run on a both-side dimension before reporting it. Omitted currency defaults to the user's Explorer currency, converting a multi-market account to that single currency exactly like the app (per-market native-currency rows appear only if no default currency is available); use account_markets to see each market's own currency. Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

read_doc

ChatGPT
Read the full text of a knowledge/ doc by its path (as returned by search_knowledge). Only paths under knowledge/ are served. Always cite the path in a 'Sources:' line.

resolve_date_range

ChatGPT
Resolve ANY relative window ('year to date', 'last 13 weeks', 'previous month', 'this quarter') to the EXACT dates the Explorer app's date picker would show — use this instead of computing dates yourself. Pass a preset (one of: this_week, previous_week, last_7_days, last_14_days, last_30_days, previous_3_months, previous_month, last_month, this_month, this_quarter, year_to_date (alias this_year), previous_year, last_13_weeks) OR the explicit triple from_type ('this'|'last'|'previous') + period (day|week|month|quarter|year) + multiplicand (default 1); the preset can also be a dynamic triple pattern like 'last_5_days', 'previous_2_weeks', etc. Today is always excluded, so day/aggregate windows end yesterday; granularity (aggregate|daily|weekly|monthly, default aggregate) snaps the range to whole weeks/months exactly like the app (a weekly year-to-date starts in the prior year and can extend past yesterday). Pass granularity = the user's defaultTimeGranularity (from my_settings) to mirror their app page — path tools consume the page's snapped range — or leave 'aggregate' for plain totals. first_day_of_week defaults to monday (the account default). Needs no account or token — pure date math. Returns {date_from, date_to, conceptual, granularity, note} (plus 'resolved' if preset was dynamically parsed).

search_knowledge

ChatGPT
Search Roivenue's knowledge base. Returns matches as {path, excerpt}; cite the paths in a 'Sources:' line. If the list is empty, answer starting with 'NO_SOURCE:'. Use this before answering any Roivenue question; never answer from outside knowledge.

submit_feedback

ChatGPT
Record feedback the user explicitly asks to pass on about this connector, in the connector's own feedback log (stored by the connector and read by Roivenue staff — the one place it goes). Use it only when the user explicitly asks to pass something on (e.g. 'tell Roivenue this was wrong', 'report this number') — never on your own initiative, and never in reaction to praise or criticism the user did not ask you to pass on. rating is 'up' or 'down'; comment carries what the user wants reported, in their words, including the specific claim or number they dispute. Include nothing the user did not ask to record. Afterwards, confirm briefly that it was recorded.

test_dimension_rule

ChatGPT
Try ONE custom-dimension rule condition against the account's real values before anything is saved — the pipeline's own rule engine runs it, nothing is written. Returns the distinct values that match (matched_values, with their weight), a sample that does not (unmatched_sample), the counts, match_share, and regex_facts for a regex (.NET: matches anywhere in the value unless anchored with ^ $, always case-insensitive, one pattern per condition, named groups (?<name>...)). A pattern the engine cannot compile is returned as its error with the offset. Pass dimension + pattern (operator defaults to regex; any rule operator works), or a full when. The rule is judged alone — in a dimension, rules run in order and the first match wins. Dates are optional: without them the tool evaluates over your Explorer window and, if that holds no rows, over the account's data coverage; a window wider than the data only slows the engine. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

transaction_history

ChatGPT
The Explorer Transaction-history page's order/conversion list for an account over [date_from, date_to]: one row per conversion — {pathConversionId, timestamp, orderStatus?, netRevenue, grossProfit?}. Customer identifiers are never returned (say so if asked); orderStatus/grossProfit exist only on OMS-enabled accounts. Sort with sort_by (pathConversionId|timestamp|orderStatus|netRevenue|grossProfit) + sort_dir; page with limit (default 20, max 50) + offset — a _truncated sentinel flags more rows. Pass a row's pathConversionId to conversion_detail for that order's touchpoint journey. attribution_model takes a model code, never a display name. An orderStatus filter matches the page's own column filter server-side — use this tool, not query_performance, for order counts. Scope with filters like [{"name": "platformName", "values": ["Google Ads"]}] (optional "operator": in|notIn|contains|notContains; contains takes one "value"). Relative windows ('last 7 days', 'year to date') exclude today and end YESTERDAY; get exact app-matching dates from resolve_date_range instead of computing them. Omit dates/attribution/currency to use the caller's Explorer defaults; the leading _scope/scope header echoes the account, dates, model (code + name), currency and dimensions actually used — report those. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

umm_setup

ChatGPT
What UMM (unified marketing measurement: MMM results mixed with MTA attribution) is set up on the account: status (not_set_up / partial / configured), the UMM and MMM models, each UMM model's per-site mixing configuration (totals source, mixing metric, period rules, MTA change constraints, and per driver the MTA/MMM weight or the target % from MMM as the UMM site settings show them; a driver marked configured: false has no settings there), the categorised MMM drivers, the experiments, and under analysis the model codes and driver-category dimension to pass to query_performance for a UMM vs MTA vs MMM comparison — UMM results are rows of the performance dataset selected by attribution model code. Whether MMM data is uploaded is not visible to a Reader and is reported as unknown. Pass account_id (from list_accounts) unless the server pins one; a pin always wins.

account_markets

Claude

account_overview

Claude

connector_health

Claude

conversion_detail

Claude

data_freshness

Claude

describe_metrics

Claude

get_skill

Claude

list_accounts

Claude

list_dimension_values

Claude

my_settings

Claude

path_histogram

Claude

path_journeys

Claude

path_kpis

Claude

path_overlap

Claude

query_performance

Claude

read_doc

Claude

resolve_date_range

Claude

search_knowledge

Claude

submit_feedback

Claude

transaction_history

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/roivenue"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/roivenue)