Hightouch MCP server

What the model can do

Hightouch reverse ETL: read sources, destinations, models, syncs with their status and each sync's runs and drafts, and run a sync, sequence or identity graph now, pause, resume or delete a sync, and approve or reject sync drafts.

Reads

• hightouch_list_sourcesHightouch sources (the warehouses and databases data is read from): id, name, type, when changed.
• hightouch_list_destinationsHightouch destinations (Salesforce, HubSpot, an ad platform…): id, name, type, how many syncs feed it, when changed.
• hightouch_list_modelsHightouch models (the queries that select rows to sync): id, name, description, source id, query type, primary key, how many syncs use it…
• hightouch_list_syncsHightouch syncs: id, slug, overall status (success, failed, warning, disabled…), whether disabled, model and destination ids, schedule type…
• hightouch_sync_runsRuns of one Hightouch sync, newest first: status, start and finish, rows in the query, rows planned, succeeded and failed (added, changed, removed)…
• hightouch_list_sync_draftsDrafts (proposed changes awaiting approval) of one Hightouch sync, oldest first: id, create or update, status…

Acts, when you ask

These change something in Hightouch. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• hightouch_trigger_syncRuns a Hightouch sync now (queued behind a run already in progress): the model's changed rows are written to the destination system (a CRM…
• hightouch_trigger_sync_sequenceRuns a Hightouch sync sequence now: every sync in it runs in order, writing to each destination.
• hightouch_trigger_identity_graphRuns a Hightouch identity resolution graph now (incrementally unless full_rerun), on the customer's warehouse compute.
• hightouch_pause_syncDisables a Hightouch sync so it stops running on its schedule (and cannot be triggered) until resumed.
• hightouch_resume_syncRe-enables a disabled Hightouch sync, so it runs on its schedule again and writes to its destination.
• hightouch_delete_syncPermanently deletes a Hightouch sync (its mappings, schedule and run history).
• hightouch_approve_sync_draftApproves a Hightouch sync draft and applies its change to the live sync (a new sync is published, or the existing one changed).
• hightouch_reject_sync_draftRejects a Hightouch sync draft: it closes without applying anything, and the requester sees it rejected.

Connect it

First connect Hightouch on Dashboard → Connections, then add Memtro to your client once. The Hightouch tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared Hightouch connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Did the Salesforce sync fail last night?
• › Which Hightouch syncs are failing?
• › How many rows failed on the HubSpot contacts sync this week?
• › Which syncs feed the Google Ads destination?
• › Rerun the Salesforce accounts sync now

Questions

Is this an official Hightouch MCP server?

No. It is Memtro's MCP server with Memtro's Hightouch connector behind it: the connector reads Hightouch through its API with the credentials you connect on Dashboard → Connections, and Hightouch publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in Hightouch?

Only when you ask, and only what it has been granted: run a sync now, sending its rows to the destination (optionally a full resync or clear and refill); run a sync sequence now; run an identity resolution graph now; pause a sync's schedule, or resume it; delete a sync permanently; approve a sync draft, applying its change, or reject it. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Databases and warehouses

• Postgres MCP server
• MySQL and MariaDB MCP server
• SQL Server MCP server
• BigQuery MCP server
• Snowflake MCP server
• AWS Athena MCP server