Skip to content
actionlake
Start free

Connectors · Developer & infrastructure

Flagsmith MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Flagsmith and the rest of your systems. It gives your agent 533 Flagsmith actions (213 reads, 238 writes and 82 that wait for a confirmation) and holds the Flagsmith credential itself, so the agent never does. It paces Flagsmith's limits and records every call.

Connect Flagsmith freeSee pricing

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: "flagsmith.organisations.list",  intent: "check which Flagsmith account is connected"}# List organisations · read · 1 upstream call
  • 533actionsin one signed file
  • 213readlook, never change
  • 238writecreate and update
  • 82wait for a yesdestructive, confirmed first

Set up

Connect Flagsmith 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 Flagsmith.

    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_keyFlagsmith Organisation API key (Organisation Settings > API Keys > Create API Key)
    • hostWhere your Flagsmith lives: your host
    account { op: "connections.requirements", service: "flagsmith" }
# 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: "flagsmith.organisations.list" }# list organisations · read · 1 upstream callcheck { action: "flagsmith.projects.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Flagsmith 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
Setup fields
host
Limits we pace to
300 calls a minute, per connection
API covered
Flagsmith Admin API v1 (OpenAPI 3.1 document served at /api/v1/swagger.json)api.flagsmith.com
Built from
Flagsmith's documentation, read 23 October 2026api.flagsmith.comdocs.flagsmith.com
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
209 of 213 reads, for up to 5 minutes; a write clears what it affects
Shared entities
4, so ask can line Flagsmith up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Flagsmith.

533 actions in 12 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 213readlook, never change
  • 238writecreate and update
  • 82destructivewait for a confirmation

  1. Environments

    209

    74 read · 102 write · 33 destructive

    Create an environment · List environments · Create a server-side API key (environment) · List server-side API keys (environment)

  2. Organisations

    146

    60 read · 60 write · 26 destructive

    List organisations · List groups (organisation) · List roles (organisation) · List users (organisation)

  3. Projects

    127

    53 read · 56 write · 18 destructive

    Create a project · List projects · List audits (project) · Create a change request (project)

  4. Features

    21

    9 read · 10 write · 2 destructive

    Create a feature export · Create a feature segment (feature) · List feature segments (feature) · Create a feature state (feature)

  5. Auth

    12

    4 read · 6 write · 2 destructive

    Create a SAML attribute mapping (auth / saml) · List SAML attribute mappings (auth / saml) · Create a SAML configuration (auth / saml) · List SAML configurations (auth / saml)

  6. Admin

    6

    6 read · 0 write · 0 destructive

    Get integrations (admin / dashboard) · Get organisations (admin / dashboard) · Get release pipelines (admin / dashboard) · Get stale flags (admin / dashboard)

  7. Metadata

    6

    2 read · 3 write · 1 destructive

    Create a metadata field (metadata) · List metadata fields (metadata) · Get a metadata field (metadata) · Replace a metadata field (metadata) (full update)

  8. Audit

    2

    2 read · 0 write · 0 destructive

    List audits · Get an audit

4 more areas · 4 actions
  • Environment feature versions 1
  • Multivariate 1
  • Segments 1
  • Webhooks 1

Top actions

The Flagsmith API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems.

ActionWhat it doesEffectSame name in
flagsmith.organisations.listList organisationsList organisations.readthis connector
flagsmith.organisations.roles.listList roles (organisation)List roles (organisation).readthis connector
flagsmith.audit.listList auditsList audits.readthis connector
flagsmith.projects.listList projectsList projects.readthis connector
flagsmith.projects.createCreate a projectCreate a project.writethis connector
flagsmith.environments.listList environmentsList environments.readthis connector
flagsmith.environments.createCreate an environmentCreate an environment.writethis connector
flagsmith.features.featurestates.listList feature states (feature)List feature states (feature).readthis connector
flagsmith.features.featurestates.createCreate a feature state (feature)Create a feature state (feature).writethis connector
flagsmith.audit.getGet an auditGet an audit.readthis connector
flagsmith.projects.deleteDelete a projectDelete a project.destructivethis connector
flagsmith.webhooks.testSend a test webhookSend a test webhook.writethis connector

Try it

Prompts to paste into Claude.

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

  1. Readread

    Who has access in Flagsmith? List every user and their role.

    callsflagsmith.organisations.users.list

  2. Writewrite

    Create a project in Flagsmith. Show me exactly what you will send before you save it.

    callsflagsmith.projects.create

Governance for Flagsmith

Destructive actions wait for a yes.

82 Flagsmith actions are destructive. The first 10:

  • Delete a projectflagsmith.projects.delete
  • Delete an environmentflagsmith.environments.delete
  • Delete an organisationflagsmith.organisations.delete
  • Delete a tag (project)flagsmith.projects.tags.delete
  • Delete a feature (project)flagsmith.projects.features.delete
  • Delete a segment (project)flagsmith.projects.segments.delete
  • Delete scim (organisation)flagsmith.organisations.scim.delete
  • Delete a trait key from every identity in an environmentflagsmith.environments.delete_traits
  • Remove users from an organisationflagsmith.organisations.remove_users
  • Delete a role (organisation)flagsmith.organisations.roles.delete

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.

doflagsmith.projects.deleteconfirm_policy: human

Delete a project?

intent: the user asked me to delete a project

run suspended · waiting for a named person · key flagsmith-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "flagsmith-agent",  allowed_actions: ["flagsmith.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "flagsmith.projects.delete", approver: "[email protected]" },    { match: "flagsmith.environments.delete", approver: "[email protected]" },    { match: "flagsmith.organisations.delete", approver: "[email protected]" }  ]}
Illustrative. Scoped to Flagsmith only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Flagsmith costs on actionlake.

Every Flagsmith action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Flagsmith 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 533 Flagsmith actions
  • 533actions pageeach page of a long list is one call
  • 209reads 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 Flagsmith action.

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

Environments209
  • flagsmith.environments.api_keys.createCreate a server-side API key (environment)write
  • flagsmith.environments.api_keys.listList server-side API keys (environment)read
  • flagsmith.environments.createCreate an environmentwrite
  • flagsmith.environments.listList environmentsread
  • flagsmith.environments.api_keys.deleteDelete a server-side API key (environment)destructive
  • flagsmith.environments.api_keys.patchUpdate a server-side API key (environment)write
  • and 203 more in this area
Organisations146
  • flagsmith.organisations.groups.listList groups (organisation)read
  • flagsmith.organisations.listList organisationsread
  • flagsmith.organisations.roles.listList roles (organisation)read
  • flagsmith.organisations.users.listList users (organisation)read
  • flagsmith.organisations.api_usage_notification.listList api usage notifications (organisation)read
  • flagsmith.organisations.audit.getGet an audit (organisation)read
  • and 140 more in this area
Projects127
  • flagsmith.projects.audit.listList audits (project)read
  • flagsmith.projects.change_requests.createCreate a change request (project)write
  • flagsmith.projects.createCreate a projectwrite
  • flagsmith.projects.listList projectsread
  • flagsmith.projects.all_user_permissions.getGet an all user permission (project)read
  • flagsmith.projects.audit.getGet an audit (project)read
  • and 121 more in this area
Features21
  • flagsmith.features.create_feature_exportCreate a feature exportwrite
  • flagsmith.features.feature_segments.createCreate a feature segment (feature)write
  • flagsmith.features.feature_segments.listList feature segments (feature)read
  • flagsmith.features.featurestates.createCreate a feature state (feature)write
  • flagsmith.features.download_feature_export.getDownload a feature exportread
  • flagsmith.features.download_flagsmith_on_flagsmith.getDownload the Flagsmith-on-Flagsmith feature exportread
  • and 15 more in this area
Auth12
  • flagsmith.auth.saml.attribute_mapping.createCreate a SAML attribute mapping (auth / saml)write
  • flagsmith.auth.saml.attribute_mapping.listList SAML attribute mappings (auth / saml)read
  • flagsmith.auth.saml.configuration.createCreate a SAML configuration (auth / saml)write
  • flagsmith.auth.saml.configuration.listList SAML configurations (auth / saml)read
  • flagsmith.auth.saml.attribute_mapping.deleteDelete a SAML attribute mapping (auth / saml)destructive
  • flagsmith.auth.saml.attribute_mapping.getGet a SAML attribute mapping (auth / saml)read
  • and 6 more in this area
Admin6
  • flagsmith.admin.dashboard.integrations.getGet integrations (admin / dashboard)read
  • flagsmith.admin.dashboard.organisations.getGet organisations (admin / dashboard)read
  • flagsmith.admin.dashboard.release_pipelines.getGet release pipelines (admin / dashboard)read
  • flagsmith.admin.dashboard.stale_flags.getGet stale flags (admin / dashboard)read
  • flagsmith.admin.dashboard.summary.getGet summary (admin / dashboard)read
  • flagsmith.admin.dashboard.usage_trends.getGet usage trends (admin / dashboard)read
Metadata6
  • flagsmith.metadata.fields.createCreate a metadata field (metadata)write
  • flagsmith.metadata.fields.getGet a metadata field (metadata)read
  • flagsmith.metadata.fields.listList metadata fields (metadata)read
  • flagsmith.metadata.fields.updateReplace a metadata field (metadata) (full update)write
  • flagsmith.metadata.fields.deleteDelete a metadata field (metadata)destructive
  • flagsmith.metadata.fields.patchUpdate a metadata field (metadata)write
Audit2
  • flagsmith.audit.getGet an auditread
  • flagsmith.audit.listList auditsread
Environment feature versions1
  • flagsmith.environment_feature_versions.getGet an environment feature versionread
Multivariate1
  • flagsmith.multivariate.options.get_by_uuid.getGet an option by UUID (multivariate)read
Segments1
  • flagsmith.segments.get_by_uuid.getGet a segment by UUIDread
Webhooks1
  • flagsmith.webhooks.testSend a test webhookwrite

This index shows a sample of each area. All 533 actions, with their inputs and prices, are in the catalogue, which a free sandbox key can search.

Questions

The Flagsmith MCP server, answered.

Is there a Flagsmith MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 533 Flagsmith actions across environments, organisations, projects and features, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Flagsmith.

What do I need to connect Flagsmith?

Flagsmith Organisation API key (Organisation Settings > API Keys > Create API Key). Plus your host. The free call account { op: "connections.requirements", service: "flagsmith" } lists exactly what to paste.

Can an AI agent delete or change things in Flagsmith?

Only if its key allows it, and only with an explicit confirmation. 82 Flagsmith actions are destructive, including delete a project, delete an environment and delete an organisation. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my Flagsmith API limit?

actionlake paces every call to the limits in the Flagsmith service file: 300 calls a minute per connection. Repeated reads are answered from cache (209 Flagsmith actions can be), which never reaches Flagsmith.

What does a Flagsmith action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (533 Flagsmith actions page). Cache hits, searches, dry runs and calls Flagsmith 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 Flagsmith API does it cover?

The Flagsmith Admin API v1 (OpenAPI 3.1 document served at /api/v1/swagger.json) at api.flagsmith.com by default, or the base URL set when you connect. It is built from Flagsmith's documentation as read on 23 October 2026. 533 actions in 12 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Flagsmith 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 Flagsmith 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.

More connectors like Flagsmith

Joins with, on shared entities

Browse

actionlake is independent. It is not made or endorsed by Flagsmith; the name is the vendor's own. Counts are from the Flagsmith service file, version 1.0.0.

Give your agent Flagsmith today.

Connect Flagsmith freeSee pricing

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