Included as the disclosed author-tested GTM context and activation platform whose official MCP gateway lets external models use the same identity-resolved layer as RevSure agents, with scoped tokens, field/module RBAC, PII redaction, immutable logging, and reversible action controls.
Concrete capability record
A concrete inventory separated by evidence type: first-party product documentation, named connector documentation, and Harry Hawk’s direct MCP tool-catalog test. Harry Hawk has a professional relationship with RevSure; this disclosed implementation context is not an endorsement or procurement recommendation.
Leads and contacts; accounts; opportunities; campaigns and spend; funnel stages and stage movement; journeys and touchpoints; meetings and engagement signals; attribution and conversion paths; pipeline and bookings; forecasts and pacing; propensity scores, personas, enrichment, and LinkedIn context; connector and synchronization state; Reverse ETL pipelines; and SQL-queryable reporting, entity, and attribution output tables.
The tested MCP catalog retrieves lead, account, opportunity, and visitor detail; entity journeys; meeting intelligence; engagement signals; funnel health and cohort movement; pipeline and booking trends; campaign metrics and attribution; conversion paths; incrementality and lift; campaign reallocation; marketing-mix contribution and response curves; buyer personas; prospects; and LinkedIn context. Separately, first-party documentation describes customer-accessible BigQuery reporting and attribution tables that can be queried by BI tools, plus Enterprise data-warehouse access including BigQuery and Snowflake. This is warehouse-mediated query access, not direct SQL access to RevSure's internal operational store.
The tested catalog includes LinkedIn connection, messaging, and post-engagement actions. Separately from MCP, RevSure documents standard CRM, MAP, and other source-system writebacks; advanced CRM writeback and Reverse ETL; and data-warehouse access. Current first-party integration guidance separately documents Snowflake writeback. Exact destinations, objects, mappings, and plan availability require implementation validation. Do not infer that platform writebacks are exposed as MCP tools in every tenant.
The MCP page documents scoped, revocable token authentication, role-based read/write scopes, PII masking, and an audit ledger. Connector documentation describes OAuth or key-based authentication with least-privilege scopes; the Claude setup guide documents OAuth client configuration.
The documented Safe Autonomy sequence is propose, approve, commit, and roll back. The MCP page says reads, writes, and decisions are logged. Current integration guides document connector-specific authentication, permissions, mappings, and synchronization; broader monitoring claims require implementation validation.
RevSure’s public MCP page describes capability and governance families but does not publish a canonical tool-name manifest. The 67-name catalog below is a tenant-specific direct-test snapshot from July 10, 2026; availability can vary by tenant, plan, configuration, role, and server release. Data-warehouse access and advanced Reverse ETL are plan and deployment dependent. Destination-specific scopes, field mappings, identity matching, cadence, deduplication, failure recovery, and whether an action is exposed through MCP, API, webhook, or a managed connector must be verified before implementation.
Direct MCP catalog observation
Observed mounted MCP catalog · captured Jul 10, 2026. The namespace was observed directly; this audit did not execute every tool. It is not a claim that public documentation publishes a permanent universal manifest.
account_detailsaccount_enrichmentaccount_journeyget_account_prospectsget_opportunity_contactslead_detailslead_enrichmentlead_journeyopportunity_detailsopportunity_journeyvisitor_detailsvisitor_journeyget_engagement_signalsget_meeting_detailsget_meetingsbooking_daily_trendscohort_movementcohort_stage_viewcohort_trenddaily_pacing_daily_breakdowndaily_pacing_stage_metricsfunnel_additionsfunnel_health_metricsopen_pipeline_trends_by_close_quarterpipeline_generation_trendspipeline_progression_trendsaggregate_journeyaggregate_journey_drilldownaggregate_journey_entitiesaggregate_touchpointaggregate_touchpoint_compositionaggregate_touchpoint_drilldowncampaign_metrics_with_attributioncampaign_metrics_without_attributionconversion_attributionconversion_pathdeep_funnel_creation_attributiondeep_funnel_leakage_attributiondrill_down_without_attributionacross_campaign_lift_analysiscampaign_reallocationconversion_lift_analysisincrementality_prepost_analysisindividual_campaign_lift_analysismmx_channel_bar_trendsmmx_channel_performance_comparisonmmx_channel_performance_trendsmmx_contribution_by_channelmmx_contribution_by_quartermmx_marketing_movementmmx_response_curve_analysisget_account_linkedin_dataget_buyer_personasget_person_linkedin_datalinkedin_connect_accountlinkedin_engage_with_postlinkedin_get_user_postslinkedin_send_connection_requestlinkedin_send_messageaccount_engagement_trendsfind_matching_dimensionsget_calendar_configget_funnel_stagesget_parameter_optionsinit_contextrecall_memoryremember_memoryFirst-party connector catalog
Connectors feed RevSure’s governed context layer. They are not automatically equivalent to individual MCP tools.
First-party access
Agent access
Support: Official MCP server
Read scope: Official MCP documentation exposes the governed context layer to external agents and states that RBAC selects readable modules and fields, with PII redaction before model access and every read logged. The public page does not publish a permanent universal tool-name manifest; the separate author-observed catalog remains a tenant-specific snapshot.
Write scope: First-party MCP material confirms role-scoped writes wrapped in propose, approve, commit, and rollback, but does not enumerate a canonical MCP mutation catalog. REST record/segment writes, CRM and warehouse writebacks, and ad-audience activation are adjacent platform surfaces and must be mapped separately rather than assumed to be MCP tools.
Authentication: Tenant-specific MCP endpoint configured with an OAuth client ID and interactive RevSure login; after authorization, the MCP gateway describes every agent call as carrying a scoped, revocable token. Claude setup leaves the client secret blank, while ChatGPT setup uses a workspace-admin-created custom app and per-user connection.
Approval boundary: Separate client invocation permission from product action approval. RevSure's MCP gateway applies a propose-approve-commit-rollback lifecycle: low-risk gates may be removed, consequential actions retain per-action approval in Slack or RevSure, and committed actions are reversible. Claude's recommended `Always Allow` setting only prevents Claude from prompting on every MCP tool call; it is not documented as bypassing RevSure's product-side Safe Autonomy gates.
Confirm the current tool catalog, plan, region, scopes, rate limits, terms, and write behavior before implementation.
Editorial assessment
A comparative architecture lens—not a quality score, market ranking, or buying recommendation. Scale: 1–5.
Proposed customer-shaped experiences
The output could be an export, report, graph, artifact, application, agent, workflow, or downstream feed. These proposals are derived from documented access—not claims that RevSure ships them.
Evidence and dates
“Publisher updated” is shown only when the page exposes an update date. “BYO-UI reviewed” records when this research checked the reference. A missing publisher date is reported as missing—not replaced with the review date.
First-party product and setup pages confirm OAuth-connected MCP access and policy controls. Public documentation does not expose a complete canonical tool manifest, so operation-level claims should carry medium…
Targeted first-party MCP/agent-role review; broader API inventory retained
REST API access with tenant and region-specific base URLs, API-token authentication, documented resources, errors, and examples.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
Named CRM, MAP, advertising, analytics, warehouse, collaboration, intent, enrichment, and engagement integrations.
Targeted first-party connector review
Snowflake writeback architecture, destination permissions, encryption, and prerequisites.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
First-party product homepage used to confirm product identity and current positioning.
Inventory source: structurally normalized; content was not individually reopened in this pass.
Claude connector OAuth setup and connection boundary.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
ChatGPT workspace administrator setup and connection boundary.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
Enterprise plan evidence for Reverse ETL and BigQuery, Snowflake, and other data-warehouse access.
Targeted first-party access review
First-party collection SDK; not an MCP capability.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
Connector authentication, key protection, network boundary, and implementation steps.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
Partner-specific enrichment request and response webhooks with documented retries and provisioned endpoints.
Full-profile first-party re-verification on 2026-07-12; see the private audit ledger.
Tied to cited first-party evidence reviewed for this profile.
Official links recorded for deeper research but not necessarily reopened endpoint by endpoint.
System role, maturity interpretation, and architectural boundary.
Interface patterns and compositions—not vendor product claims.