Skip to main content

MoltSets Quick Start Guide

Everything you need to make your first MoltSets API call, from creating an API key to getting a real response back.

MoltSets can be connected to your workflow in two ways: through our API or through our MCP server. Which one you use depends on what you're building and how technical you want to get.


Path 1: Connect via API

Who this is for

Developers and technical users who want full control, plan to build custom integrations, or need to embed MoltSets data into their own systems, scripts, or products.

Use case

You're writing code that calls MoltSets directly, whether that's a one-off script, a production integration, or a feature inside your own app.

MoltSets API Setup Instructions

Getting started

  1. Go to API Keys to generate your API key.

  2. Head to our developer docs at https://developer.moltsets.com to see full endpoint documentation.

Each API has a playground built into the docs, so you can test requests and see live responses before writing any code.


Path 2: Connect via MCP

Who this is for

Users who want to work with MoltSets conversationally inside Claude, without writing code. This is the faster path if you're using Claude for prospecting, enrichment, or research and want MoltSets tools available directly in chat.

Use case

You want Claude to call MoltSets tools (lookups, enrichment, search) as part of a normal conversation, agentic workflow, or Claude Code session, instead of hitting the API yourself.

Before you start

Before you start, you'll need your MoltSets Connector credentials (OAuth Client ID and Client Secret). Get these by logging into app.moltsets.com and going to API Keys to create a new connector.

The setup steps depend on which Claude client, or other MCP client, you're using.

MoltSets MCP Setup Instructions

If you're using Claude.ai — Personal account (Free, Pro, or Max)

  1. Go to Customize → Connectors.

  2. Click +, then Add custom connector.

  3. Paste this into Remote MCP server URL:

    https://mcp.moltsets.com/mcp
  4. Click Advanced settings and enter your OAuth Client ID and Client Secret.

  5. Click Add.

Free accounts are limited to a single custom connector. Pro and Max accounts can add multiple.


If you're using Claude.ai — Team or Enterprise account

Only an Owner or Primary Owner of your Claude Team/Enterprise account can add a custom connector for the organization. Once it's added, each member connects individually, and Claude only ever accesses data that member is authorized to see.

Before you begin

You will first need to invite your Owner or Primary Owner of your Claude Team/Enterprise account to MoltSets as they will need to have a MoltSets user in order to complete the connection process.

Owner or Primary Owner — add the connector:

  1. Go to Organization settings → Connectors.

  2. Click Add.

  3. Hover over Custom, then select Web.

  4. Paste this into Remote MCP server URL:

    https://mcp.moltsets.com/mcp
  5. Click Advanced settings and enter the OAuth Client ID and Client Secret.

  6. Click Add.

Each member — connect to it:

  1. Go to Customize → Connectors.

  2. Find the MoltSets connector (marked Custom).

  3. Click Connect and complete authentication.


If you're using Claude Desktop

Claude Desktop shares the same connectors as your Claude.ai account, so if you've already added MoltSets there, no further setup is needed here.

Personal account (Free, Pro, or Max):

  1. Go to Customize → Connectors.

  2. Click +, then Add custom connector.

  3. Paste this into Remote MCP server URL:

    https://mcp.moltsets.com/mcp
  4. Click Advanced settings and enter your OAuth Client ID and Client Secret.

  5. Click Add, then restart Claude Desktop so it picks up the connector.

Team or Enterprise account:

Same organization-level setup as Claude.ai above (an Owner adds the connector once), then each member goes to Customize → Connectors in Claude Desktop, finds MoltSets, clicks Connect, and restarts the app.

Tool permissions:

By default, all MoltSets tools are set to Needs Approval, so Claude will prompt before using each one the first time. To change this in bulk, go to Customize → Connectors, click Configure beside MoltSets, and use the bulk change tool to set tools to Always Allow. Individual tools can also be set to Blocked if you want to restrict access.


If you're using Claude Code

Run this in your terminal, replacing the placeholders with your credentials:

claude mcp add --transport http moltsets https://mcp.moltsets.com/mcp \   --client-id YOUR_CLIENT_ID \   --client-secret

The --client-secret flag takes no value. Claude Code will prompt you for it, or you can set it via the MCP_CLIENT_SECRET environment variable.

By default this adds MoltSets at the local scope (current project only). Add --scope user to make it available across all your projects:

claude mcp add --transport http moltsets https://mcp.moltsets.com/mcp \   --scope user \   --client-id YOUR_CLIENT_ID \   --client-secret

Claude Code will open a browser window for you to approve access on the MoltSets consent screen. Once approved, your token is stored and managed automatically.

Personal account:

The command above works as-is.

Team or Enterprise account:

Only admins can add MCP servers to Claude Code on business plans. If you sign into Claude Code with your Claude.ai account, any MoltSets connector already added there is available automatically, no claude mcp add needed. Either way each seat authorizes over OAuth in the browser, so there's no need to distribute ms_ API keys across an organization.


If you're using another MCP client

Any MCP client that supports remote servers over HTTP can connect — CLI agents, desktop apps, and IDE extensions included.

  1. Add https://mcp.moltsets.com/mcp as a remote MCP server in your client.

  2. When the client asks for OAuth credentials, paste the Client ID and Client Secret from the API Keys page at app.moltsets.com. Clients that register themselves can skip this step.

  3. Approve access on the MoltSets consent screen when the browser opens. The client stores the token from there; re-run the flow if it expires.

Clients that run in a terminal complete this on a local port — http://127.0.0.1:<port>/callback and its localhost and [::1] equivalents are supported, so no HTTPS callback is needed on your machine.

Endpoints, scopes, and redirect URI rules are discoverable directly from the server at https://mcp.moltsets.com/.well-known/oauth-protected-resource.


Example Prompts

Once connected, you can talk to MoltSets the same way you'd talk to a colleague. Here are a few prompts to get started:

Enrich a contact from LinkedIn

"Enrich this LinkedIn profile and give me their business email and phone number: linkedin.com/in/example"

Look up a visitor

"I have an IP address from a site visitor — 192.0.2.1 — can you tell me who this is?"

Resolve a hashed email

"Resolve this HEM to a real identity and return the name and company: abc123..."

Bulk enrich a list

"Here are 10 LinkedIn URLs. Enrich each one for business email and return the results as a table."


Reading an Outcome

Every MoltSets tool call ends in one of three outcomes, and Claude should tell them apart before it reports back to you:

Outcome

MCP response

What it means

Found

isError: false, status: "ok"

Data was found and returned. Tokens charged.

Not found

isError: false, status: "not_found"

The tool ran fine; MoltSets holds no data for that input. Nothing charged.

Error

isError: true with an error code

The call did not complete — bad input, no tokens, plan restriction, rate limit, or an outage.

A not-found is a normal, uncharged result, not a failure — the lookup was clean, the data simply isn't on file. It still counts as one request against your Fair Use requests limit.

When it really is an error

Code

Means

Next step

invalid_input

A parameter is wrong

Fix the parameter — rerunning it unchanged repeats the error

insufficient_tokens

Out of tokens

Top up or upgrade; get_billing shows the balance

insufficient_phone_tokens

Out of phone tokens

A separate allowance from regular tokens

plan_restricted

The tool isn't on your plan

A plan condition, not a data miss — switch tools or contact support

fair_use_limit_exceeded / fair_use_requests_exceeded

A Fair Use window is exhausted

Wait the number of seconds in metadata.retry_after, then resume

rate_limited

Too many calls too quickly

Back off briefly

unauthorized / account_inactive

Credentials or account problem

Recheck the connector; contact support if the account is inactive

tool_not_found

Unknown or retired tool name

Check the name against the tool list

internal_error

The endpoint is temporarily offline

Stop retrying that tool for now; nothing is charged and the team is alerted automatically

Nothing is charged on an error.

Did this answer your question?