Skip to content

Connectors · Work & projects

Lattice MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Lattice and the rest of your systems. It gives your agent 34 Lattice actions (31 reads and 3 writes) and holds the Lattice credential itself, so the agent never does. It paces Lattice'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/mcp
the first call
do {  action: "lattice.users.list",  intent: "check which Lattice account is connected"}# Returns a paginated list of users in Lattice · read · 1 upstream call
  • 34actionsin one signed file
  • 31readlook, never change
  • 3writecreate and update
  • 0wait for a yesdestructive, confirmed first

Set up

Connect Lattice in three steps.

  1. 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
  2. 02

    Connect Lattice.

    Bearer token. 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_keyLattice API key
    • hostWhere your Lattice lives: your host
    account { op: "connections.requirements", service: "lattice" }
    # free: exactly what to paste, and what is found for you
  3. 03

    Make the first call. Check before a write.

    Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

    do    { action: "lattice.users.list" }# returns a paginated list of users in lattice · read · 1 upstream callcheck { action: "lattice.goals.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Lattice at a glance

Sign-in
Bearer token. 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
Setup fields
host
Limits we pace to
240 calls a minute, per connection
API covered
Lattice Talent API v1 (OpenAPI 3.0 per reference page, developers.lattice.com)api.latticehq.com
Built from
Lattice's documentation, read 23 October 2026developers.lattice.comdevelopers.lattice.com
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
31 of 31 reads, for up to 5 minutes; a write clears what it affects
Shared entities
3, so ask can line Lattice up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Lattice.

34 actions in 16 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 31readlook, never change
  • 3writecreate and update
  • 0destructivewait for a confirmation
  1. Users

    6

    6 read · 0 write · 0 destructive

    Returns a paginated list of users in Lattice · Returns a user with the given id · Returns a list of the user's custom attributes · Returns a list of users that report to a user

  2. Goals

    4

    2 read · 2 write · 0 destructive

    Creates a new goal with the provided information · Returns a paginated list of goals in Lattice · Returns a goal with the given id · Updates an existing goal with the provided information

  3. Review cycles

    4

    4 read · 0 write · 0 destructive

    Returns a paginated list of all of the reviews for a review cycle · Returns a paginated list of all review cycles in Lattice · Returns a paginated list of all reviewees in a review cycle · Returns a review cycle with the given id

  4. Goal updates

    3

    2 read · 1 write · 0 destructive

    Creates a progress update for an existing goal · Retrieves all goal updates across all goals in the company with pagination support · Retrieves all progress updates for a specific goal with pagination support

  5. Departments

    2

    2 read · 0 write · 0 destructive

    Returns a paginated list of all departments in Lattice · Retuns a department with a given id

  6. Feedbacks

    2

    2 read · 0 write · 0 destructive

    Returns a paginated list of all continuous feedback in Lattice · Returns a feedback with the given id

  7. Reviewees

    2

    2 read · 0 write · 0 destructive

    Returns a paginated list of all of the reviews for a reviewee · Returns a reviewee with the given id

  8. Tags

    2

    2 read · 0 write · 0 destructive

    Returns a paginated list of all tags in Lattice · Retuns a tag with a given id

8 more areas · 9 actions
  • Updates 2
  • Competencies 1
  • Custom attribute values 1
  • Custom attributes 1
  • Me 1
  • Question revisions 1
  • Questions 1
  • Tasks 1

Top actions

The Lattice API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 10 of Lattice's 10 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
lattice.users.listReturns a paginated list of users in LatticeReturns a paginated list of users in Lattice.readidentity.user.list474 connectors
lattice.departments.listReturns a paginated list of all departments in LatticeReturns a paginated list of all departments in Lattice.readhr.department.list40 connectors
lattice.users.getReturns a user with the given idReturns a user with the given id.readidentity.user.get399 connectors
lattice.departments.getRetuns a department with a given idRetuns a department with a given id.readhr.department.get33 connectors
lattice.me.getReturns the current user the API token is associated withReturns the current user the API token is associated with.readidentity.current_user.get307 connectors
lattice.goals.listReturns a paginated list of goals in LatticeReturns a paginated list of goals in Lattice.readwork.goal.list7 connectors
lattice.goals.createCreates a new goal with the provided informationCreates a new goal with the provided information.writework.goal.create7 connectors
lattice.goals.updateUpdates an existing goal with the provided informationUpdates an existing goal with the provided information.writework.goal.update7 connectors
lattice.review_cycles.list_reviewsReturns a paginated list of all of the reviews for a review cycleReturns a paginated list of all of the reviews for a review cycle.readhr.performance_review.list6 connectors
lattice.updates.listReturns a paginated list of all Updates in LatticeReturns a paginated list of all Updates in Lattice.readthis connector
lattice.feedbacks.listReturns a paginated list of all continuous feedback in LatticeReturns a paginated list of all continuous feedback in Lattice.readthis connector
lattice.goal_updates.createCreates a progress update for an existing goalCreates a progress update for an existing goal.writethis connector

Try it

Prompts to paste into Claude.

Each one names the Lattice actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    List everyone in Lattice with their department and start date.

    callslattice.users.list

  2. Writewrite

    Creates a new goal with the provided information in Lattice. Show me exactly what you will send before you save it.

    callslattice.goals.create

Governance for Lattice

Every write can wait for a yes.

None of Lattice'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.

a key for this connector
account {  op: "keys.create", name: "lattice-agent",  allowed_actions: ["lattice.*"], cap_calls: 2000,  confirm_policy: "human_for_writes"}
Illustrative. Scoped to Lattice only, capped at 2,000 upstream calls, every write confirmed.

Cost

What Lattice costs on actionlake.

Every Lattice action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Lattice 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 34 Lattice actions
  • 34actions pageeach page of a long list is one call
  • 31reads can be cacheda cache hit costs nothing
  • 100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Lattice action.

34 actions by area: the id your agent calls, what it does, and its effect.

Users6
  • lattice.users.getReturns a user with the given idread
  • lattice.users.listReturns a paginated list of users in Latticeread
  • lattice.users.list_custom_attributesReturns a list of the user's custom attributesread
  • lattice.users.list_direct_reportsReturns a list of users that report to a userread
  • lattice.users.list_goalsReturns a paginated list of all of the goals owned by a userread
  • lattice.users.list_tasksReturns a paginated list of tasks in Latticeread
Goals4
  • lattice.goals.createCreates a new goal with the provided informationwrite
  • lattice.goals.getReturns a goal with the given idread
  • lattice.goals.listReturns a paginated list of goals in Latticeread
  • lattice.goals.updateUpdates an existing goal with the provided informationwrite
Review cycles4
  • lattice.review_cycles.getReturns a review cycle with the given idread
  • lattice.review_cycles.listReturns a paginated list of all review cycles in Latticeread
  • lattice.review_cycles.list_revieweesReturns a paginated list of all reviewees in a review cycleread
  • lattice.review_cycles.list_reviewsReturns a paginated list of all of the reviews for a review cycleread
Goal updates3
  • lattice.goal_updates.createCreates a progress update for an existing goalwrite
  • lattice.goal_updates.listRetrieves all goal updates across all goals in the company with pagination supportread
  • lattice.goal_updates.list_for_goalRetrieves all progress updates for a specific goal with pagination supportread
Departments2
  • lattice.departments.getRetuns a department with a given idread
  • lattice.departments.listReturns a paginated list of all departments in Latticeread
Feedbacks2
  • lattice.feedbacks.getReturns a feedback with the given idread
  • lattice.feedbacks.listReturns a paginated list of all continuous feedback in Latticeread
Reviewees2
  • lattice.reviewees.getReturns a reviewee with the given idread
  • lattice.reviewees.list_reviewsReturns a paginated list of all of the reviews for a revieweeread
Tags2
  • lattice.tags.getRetuns a tag with a given idread
  • lattice.tags.listReturns a paginated list of all tags in Latticeread
Updates2
  • lattice.updates.getReturns an Update with the given idread
  • lattice.updates.listReturns a paginated list of all Updates in Latticeread
Competencies1
  • lattice.competencies.getReturns a competency with the given idread
Custom attribute values1
  • lattice.custom_attribute_values.getReturns the custom attribute value with the given idread
Custom attributes1
  • lattice.custom_attributes.getReturns the custom attribute with the given idread
Me1
  • lattice.me.getReturns the current user the API token is associated withread
Question revisions1
  • lattice.question_revisions.getReturns a question revision with the given idread
Questions1
  • lattice.questions.getReturns a review question with the given idread
Tasks1
  • lattice.tasks.getRetuns a task with a given idread

Questions

The Lattice MCP server, answered.

Is there a Lattice MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 34 Lattice actions across users, goals, review cycles and goal updates, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Lattice.

What do I need to connect Lattice?

Lattice API key. Plus your host. The free call account { op: "connections.requirements", service: "lattice" } lists exactly what to paste.

Can an AI agent change things in Lattice?

Only if its key allows it. 3 Lattice 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 Lattice API limit?

actionlake paces every call to the limits in the Lattice service file: 240 calls a minute per connection. Repeated reads are answered from cache (31 Lattice actions can be), which never reaches Lattice.

What does a Lattice action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (34 Lattice actions page). Cache hits, searches, dry runs and calls Lattice 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 Lattice API does it cover?

The Lattice Talent API v1 (OpenAPI 3.0 per reference page, developers.lattice.com) at api.latticehq.com by default, or the base URL set when you connect. It is built from Lattice's documentation as read on 23 October 2026. 34 actions in 16 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Lattice 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 Lattice 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 Lattice today.

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.