Helicone MCP server

What the model can do

Helicone LLM observability: the requests applications sent through it with model, tokens, cost, latency, status, ratings and scores (never prompts or completions), cost by model over a window, sessions, a named end user's usage, property names, managed prompts and their versions, and datasets. Can rate, score and tag requests and sessions, create and version managed prompts and point an environment at a version, and build datasets from logged requests.

Reads

• helicone_list_requestsLLM requests logged by Helicone, newest first: id, time, model, provider, HTTP status, tokens, cost in USD, latency, time to first token…
• helicone_cost_by_modelHelicone cost in USD, request count, tokens, errors and average latency per model over a window, from the logged requests (at most 500 read…
• helicone_sessionsHelicone sessions (multi-step agent or conversation traces) over a window: id, name, first and latest request, request count, cost…
• helicone_user_usageRequests, tokens and cost in USD over a window for the end users named (the Helicone-User-Id the application sends, often an email).
• helicone_propertiesThe custom property names the company's applications attach to Helicone requests (Helicone-Property-* headers such as environment or feature)…
• helicone_list_promptsHelicone managed prompts: id, name, tags and when created.
• helicone_list_prompt_versionsVersions of one Helicone managed prompt: version id, number (major.minor), model, commit message, which environments use it and when made.
• helicone_list_datasetsHelicone datasets (sets of logged requests kept for evaluation or fine-tuning): id, name, type, how many requests and when created.

Acts, when you ask

These change something in Helicone. 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.

• helicone_rate_requestRates one Helicone request thumbs up (true) or down (false).
• helicone_score_requestAdds named scores (numbers or true/false) to one Helicone request, such as an evaluation result.
• helicone_set_request_propertySets a custom property (a label such as environment=prod or customer_tier=gold) on one logged Helicone request…
• helicone_rate_sessionRates one Helicone session (a multi-step trace) thumbs up (true) or down (false); a later rating replaces it.
• helicone_tag_sessionSets the tag of one Helicone session, replacing the tag it has (a session has one).
• helicone_create_promptCreates a Helicone managed prompt with its first version (a model and messages).
• helicone_update_promptSaves a new version of a Helicone managed prompt (a minor version, or a major one when asked), based on the version named.
• helicone_set_prompt_environmentPoints an environment (such as production or staging) of a Helicone managed prompt at the version named.
• helicone_rename_promptRenames a Helicone managed prompt.
• helicone_tag_promptSets a Helicone managed prompt's tags to exactly the list given, replacing its current tags (an empty list removes them all).
• helicone_delete_promptDeletes a Helicone managed prompt and all its versions, permanently.
• helicone_create_datasetCreates a Helicone dataset from named logged requests (1 to 100 request ids).
• helicone_change_dataset_requestsAdds and/or removes named logged requests (at most 100 of each) in a Helicone dataset.
• helicone_delete_datasetDeletes a Helicone dataset, permanently.

Connect it

First connect Helicone on Dashboard → Connections, then add Memtro to your client once. The Helicone 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 Helicone 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

• › What did we spend on each model in the last 7 days?
• › Which Helicone requests failed in the last 24 hours?
• › How much did priya@acme.com cost us in the last 30 days?
• › Which sessions cost the most this week?
• › Point production for the support-reply prompt at version 4.

Questions

Is this an official Helicone MCP server?

No. It is Memtro's MCP server with Memtro's Helicone connector behind it: the connector reads Helicone through its API with the credentials you connect on Dashboard → Connections, and Helicone 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 Helicone?

Only when you ask, and only what it has been granted: rate a Helicone request thumbs up or down, when it has no rating yet; add named scores to a Helicone request, when it has no score of that name yet; set a custom property on one Helicone request, replacing any value of that key; rate a Helicone session thumbs up or down; set a Helicone session's tag, replacing the one it has; create a Helicone managed prompt with its first version; save a new version of a Helicone managed prompt (not deployed to any environment); point a Helicone environment such as production at a prompt version, which live traffic then uses; rename a Helicone managed prompt; replace a Helicone managed prompt's tags; delete a Helicone managed prompt and its versions permanently; create a Helicone dataset from named logged requests; add or remove named requests in a Helicone dataset; delete a Helicone dataset permanently. 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 · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server