Connectors · CRM & sales
Clay MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Clay and the rest of your systems. It gives your agent 16 Clay actions (11 reads and 5 writes) and holds the Clay credential itself, so the agent never does. It paces Clay's limits and records every call.
100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.
claude mcp add --transport http actionlake https://connect.actionlake.com/mcpdo { action: "clay.me.get", intent: "check which Clay account is connected"}# Get the authenticated user and workspace · read · 1 upstream call- 16actionsin one signed file
- 11readlook, never change
- 5writecreate and update
- 0wait for a yesdestructive, confirmed first
Set up
Connect Clay in three steps.
- 01
Add actionlake to Claude.
Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
https://connect.actionlake.com/mcp - 02
Connect Clay.
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
api_keyClay API key: app.clay.com > API and CLI > API keys (or `clay api-keys create`)
account { op: "connections.requirements", service: "clay" } # free: exactly what to paste, and what is found for you - 03
Make the first call. Check before a write.
Ask in plain words; Claude picks the action. Before anything changes,
checkvalidates the call and prices it, for free.do { action: "clay.me.get" }# get the authenticated user and workspace · read · 1 upstream callcheck { action: "clay.searches.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Clay at a glance
- Sign-in
- API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
- You provide
api_key- Limits we pace to
- 120 calls a minute, per connection
- API covered
- Clay Public API v0 (OpenAPI 3.1, developers.clay.com/openapi.json)
api.clay.com/public/v0 - Built from
- Clay's documentation, read 23 October 2026developers.clay.comdevelopers.clay.com
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 7 of 11 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 1, so ask can line Clay up with your other systems
- Data
- Processed and stored in Sydney, Australia. No US region yet
Capabilities
What an AI agent can do in Clay.
16 actions in 12 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 11readlook, never change
- 5writecreate and update
- 0destructivewait for a confirmation
Filter searches
21 read · 1 write · 0 destructive
Create a search from structured filters (deprecated) · Fetch the next page of a filters-mode search (deprecated)
Routine batch runs
21 read · 1 write · 0 destructive
Start a batch routine run over an uploaded JSONL file · Get progress and results of a batch routine run
Routine runs
21 read · 1 write · 0 destructive
Run a routine (function) against 1-100 items · Get progress and results of a routine run
Searches
21 read · 1 write · 0 destructive
Create a search from a Clay search query · Fetch the next page of a search
Credits
11 read · 0 write · 0 destructive
Get workspace credit balances
Me
11 read · 0 write · 0 destructive
Get the authenticated user and workspace
Routine batch uploads
10 read · 1 write · 0 destructive
Get a presigned upload URL for a batch input file
Search filter fields
11 read · 0 write · 0 destructive
List filter fields for a filters-mode search (deprecated)
4 more areas · 4 actions
- Search query reference 1
- Tables 1
- Workflow run query reference 1
- Workflow runs 1
Top actions
The Clay API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 1 of Clay's 1 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
clay.me.getGet the authenticated user and workspace | Returns the Clay user and workspace the API key belongs to (a free call: use it to test a connection). | read | identity.current_user.get307 connectors |
clay.searches.runFetch the next page of a search | Run the query-mode iterator and return the next page of results. | read | this connector |
clay.tables.queryQuery Clay tables | Run a structured query across one or more tables. | read | this connector |
clay.searches.createCreate a search from a Clay search query | Create a search from a Clay search query. | write | this connector |
clay.routine_runs.createRun a routine (function) against 1-100 items | Execute a routine against 1-100 items. | write | this connector |
clay.credits.get_balanceGet workspace credit balances | Returns the workspace's current data-credit and action-execution balances; reading balances costs nothing. | read | this connector |
clay.routine_runs.get_resultsGet progress and results of a routine run | Fetch progress and results for a routine run. | read | this connector |
clay.filter_searches.runFetch the next page of a filters-mode search (deprecated) | Run the search iterator and return the next page of results. | read | this connector |
clay.workflow_runs.searchSearch workflow runs by query (beta) | Search runs by query (beta). | read | this connector |
clay.filter_searches.createCreate a search from structured filters (deprecated) | Create a search from structured filters. | write | this connector |
clay.routine_batch_runs.createStart a batch routine run over an uploaded JSONL file | Start an async routine run-batch over an uploaded JSONL file. | write | this connector |
clay.search_filter_fields.listList filter fields for a filters-mode search (deprecated) | List the filter fields available for a search source type. | read | this connector |
Try it
Prompts to paste into Claude.
Each one names the Clay actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Get the authenticated user and workspace in Clay and tell me what stands out.
calls
clay.me.get - Writewrite
Create a search from a Clay search query in Clay. Show me exactly what you will send before you save it.
calls
clay.searches.create
Governance for Clay
Every write can wait for a yes.
None of Clay's actions is destructive. A key can still be held to reads, capped, and set so a person confirms every write.
A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.
account { op: "keys.create", name: "clay-agent", allowed_actions: ["clay.*"], cap_calls: 2000, confirm_policy: "human_for_writes"}Cost
What Clay costs on actionlake.
Every Clay action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Clay refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
- 1upstream call per actionall 16 Clay actions
- 16actions pageeach page of a long list is one call
- 7reads can be cacheda cache hit costs nothing
- 100,000calls a month, freepaid plans from US$49 a month for 250,000
Action index
Every Clay action.
16 actions by area: the id your agent calls, what it does, and its effect.
Filter searches2
clay.filter_searches.createCreate a search from structured filters (deprecated)writeclay.filter_searches.runFetch the next page of a filters-mode search (deprecated)read
Routine batch runs2
clay.routine_batch_runs.createStart a batch routine run over an uploaded JSONL filewriteclay.routine_batch_runs.get_resultsGet progress and results of a batch routine runread
Routine runs2
clay.routine_runs.createRun a routine (function) against 1-100 itemswriteclay.routine_runs.get_resultsGet progress and results of a routine runread
Searches2
clay.searches.createCreate a search from a Clay search querywriteclay.searches.runFetch the next page of a searchread
Credits1
clay.credits.get_balanceGet workspace credit balancesread
Me1
clay.me.getGet the authenticated user and workspaceread
Routine batch uploads1
clay.routine_batch_uploads.createGet a presigned upload URL for a batch input filewrite
Search filter fields1
clay.search_filter_fields.listList filter fields for a filters-mode search (deprecated)read
Search query reference1
clay.search_query_reference.getGet the Clay search query referenceread
Tables1
clay.tables.queryQuery Clay tablesread
Workflow run query reference1
clay.workflow_run_query_reference.getGet the workflow-run query reference (beta)read
Workflow runs1
clay.workflow_runs.searchSearch workflow runs by query (beta)read
Questions
The Clay MCP server, answered.
Is there a Clay MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 16 Clay actions across filter searches, routine batch runs, routine runs and searches, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Clay.
What do I need to connect Clay?
Clay API key: app.clay.com > API and CLI > API keys (or `clay api-keys create`). The free call account { op: "connections.requirements", service: "clay" } lists exactly what to paste.
Can an AI agent change things in Clay?
Only if its key allows it. 5 Clay actions write, and none is destructive. A key can be limited to reads, capped, and set so a person confirms every write.
Will an AI agent use up my Clay API limit?
actionlake paces every call to the limits in the Clay service file: 120 calls a minute per connection. Repeated reads are answered from cache (7 Clay actions can be), which never reaches Clay.
What does a Clay action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (16 Clay actions page). Cache hits, searches, dry runs and calls Clay refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
Which Clay API does it cover?
The Clay Public API v0 (OpenAPI 3.1, developers.clay.com/openapi.json) at api.clay.com/public/v0. It is built from Clay's documentation as read on 23 October 2026. 16 actions in 12 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Clay data, and where?
actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Clay credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.
Which AI clients can use it?
Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.
Give your agent Clay today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.
