Skip to content

Connectors · Developer & infrastructure

LaunchDarkly MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to LaunchDarkly and the rest of your systems. It gives your agent 401 LaunchDarkly actions (200 reads, 149 writes and 52 that wait for a confirmation) and holds the LaunchDarkly credential itself, so the agent never does. It paces LaunchDarkly'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: "launchdarkly.members.list",  intent: "check which LaunchDarkly account is connected"}# List account members · read · 1 upstream call
  • 401actionsin one signed file
  • 200readlook, never change
  • 149writecreate and update
  • 52wait for a yesdestructive, confirmed first

Set up

Connect LaunchDarkly 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 LaunchDarkly.

    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.

    • access_tokenLaunchDarkly API access token (personal or service token)
    • hostWhere your LaunchDarkly lives: your host
    account { op: "connections.requirements", service: "launchdarkly" }
    # 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: "launchdarkly.members.list" }# list account members · read · 1 upstream callcheck { action: "launchdarkly.teams.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

LaunchDarkly 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
access_token
Setup fields
host
Limits we pace to
10 calls a second, per connection
API covered
LaunchDarkly REST API v2 (OpenAPI 3.0.3)app.launchdarkly.com
Built from
LaunchDarkly's documentation, read 22 October 2026app.launchdarkly.comlaunchdarkly.com
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
168 of 200 reads, for up to 5 minutes; a write clears what it affects
Shared entities
5, so ask can line LaunchDarkly up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in LaunchDarkly.

401 actions in 146 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 200readlook, never change
  • 149writecreate and update
  • 52destructivewait for a confirmation
  1. Flag approval requests

    8

    2 read · 5 write · 1 destructive

    Apply approval request for a flag · Create approval request for a flag · Create approval request to copy flag configurations across environments · List approval requests for a flag

  2. Adaptive triggers

    7

    2 read · 4 write · 1 destructive

    Create an adaptive trigger · Disable an adaptive trigger · Enable an adaptive trigger · List adaptive triggers

  3. Approval requests

    7

    2 read · 4 write · 1 destructive

    Apply approval request · Create approval request · List approval requests · Review approval request

  4. Environments

    7

    2 read · 4 write · 1 destructive

    List environments · Create environment · Get environment · Update environment

  5. Feature flags

    7

    2 read · 4 write · 1 destructive

    List feature flags · Create a feature flag · Get feature flag · Update feature flag

  6. Integration delivery configurations

    7

    3 read · 3 write · 1 destructive

    Create delivery configuration · List all delivery configurations · Get delivery configurations by environment · Validate delivery configuration

  7. Flag import configurations

    6

    2 read · 3 write · 1 destructive

    Create a flag import configuration · List all flag import configurations · Trigger a single flag import run · Get a single flag import configuration

  8. Members

    6

    2 read · 3 write · 1 destructive

    List account members · Get account member · Modify an account member · Delete account member

138 more areas · 346 actions
  • Relay proxy configs 6
  • Release policies 6
  • SDK keys 6
  • Segments 6
  • Teams 6
  • Tokens 6
  • Agent graphs 5
  • Agent optimization results 5
  • Agent optimizations 5
  • Agent skills 5
  • AI configs 5
  • AI tools 5
  • Big segment store integrations 5
  • Code ref repositories 5
  • Custom roles 5
  • Destinations 5
  • Experiments 5
  • Flag scheduled changes 5
  • Flag triggers 5
  • Holdouts 5
  • Insights charts 5
  • Insights groups 5
  • Integration configurations 5
  • Integration subscriptions 5
  • Metric groups 5
  • Metrics 5
  • Model configs 5
  • Oauth clients 5
  • Projects 5
  • Prompt snippets 5
  • Release pipelines 5
  • Views 5
  • Warehouse destinations 5
  • Webhooks 5
  • AI config variations 4
  • Announcements 4
  • Applications 4
  • Code ref branches 4
  • Flag followers 4
  • Flag links 4
  • Flag releases 4
  • Flag workflows 4
  • Application versions 3
  • Audit log entries 3
  • Feature flag statuses 3
  • Insights deployments 3
  • Ip allowlist entries 3
  • Layers 3
  • Project flag defaults 3
  • User flag settings 3
  • Users 3
  • View linked resources 3
  • Workflow templates 3
  • Agent optimization runs 2
  • AI config metrics 2
  • AI config targeting 2
  • AI restricted models 2
  • Approval request settings 2
  • Big segment exports 2
  • Code ref extinctions 2
  • Code ref statistics 2
  • Context attributes 2
  • Context kinds 2
  • Contexts 2
  • Dependent flags 2
  • Experimentation settings 2
  • Flag expiring context targets 2
  • Flag expiring user targets 2
  • Insights repository projects 2
  • Ip allowlist 2
  • Segment expiring context targets 2
  • Segment expiring user targets 2
  • User expiring flags 2
  • Agent skill references 1
  • Agent skill versions 1
  • AI config quick stats 1
  • AI runs usage 1
  • AI tool references 1
  • AI tool versions 1
  • API versions 1
  • Audit log entry counts 1
  • Big segment context targets 1
  • Big segment imports 1
  • Big segment user targets 1
  • Caller identity 1
  • Context flag settings 1
  • Context instance segment memberships 1
  • Context instances 1
  • Contexts clientside usage 1
  • Contexts serverside usage 1
  • Contexts total usage 1
  • Data export events usage 1
  • Data export project rankings 1
  • Evaluations usage 1
  • Events usage 1
  • Experiment iterations 1
  • Experimentation events usage 1
  • Experimentation keys usage 1
  • Flag release phases 1
  • Insights deployment events 1
  • Insights flag events 1
  • Insights pull requests 1
  • Insights repositories 1
  • Insights scores 1
  • Linked views 1
  • Mau clientside usage 1
  • Mau SDKS by types 1
  • Mau total usage 1
  • Mau usage 1
  • Mau usage by categories 1
  • Member teams 1
  • Migration safety issues 1
  • Model config versions 1
  • Observability errors usage 1
  • Observability logs usage 1
  • Observability metrics usage 1
  • Observability sessions usage 1
  • Observability traces usage 1
  • Openapi spec 1
  • Prompt snippet references 1
  • Prompt snippet versions 1
  • Public ip list 1
  • Release pipeline progressions 1
  • Root 1
  • SDK all versions 1
  • SDK key views 1
  • SDK versions details 1
  • Segment context memberships 1
  • Segment user memberships 1
  • Service connections usage 1
  • Stream usage 1
  • Stream usage by SDK versions 1
  • Stream usage sdkversions 1
  • Tags 1
  • Team maintainers 1
  • Team roles 1
  • Vega AI usage 1
  • Warehouse export usage 1

Top actions

The LaunchDarkly API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 60 of LaunchDarkly's 61 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
launchdarkly.members.listList account membersList account members.readidentity.user.list474 connectors
launchdarkly.teams.listList teamsList teams.readidentity.group.list262 connectors
launchdarkly.custom_roles.listList custom rolesList custom roles.readidentity.role.list184 connectors
launchdarkly.teams.createCreate teamCreate team.writeidentity.group.create205 connectors
launchdarkly.members.getGet account memberGet account member.readidentity.user.get399 connectors
launchdarkly.custom_roles.createCreate custom roleCreate custom role.writeidentity.role.create105 connectors
launchdarkly.members.deleteDelete account memberDelete account member.destructiveidentity.user.delete253 connectors
launchdarkly.code_ref_repositories.listList repositoriesList repositories.readdevops.repository.list28 connectors
launchdarkly.projects.listList projectsList projects.readwork.project.list109 connectors
launchdarkly.tokens.createCreate access tokenCreate access token.writeidentity.api_key.create100 connectors
launchdarkly.projects.createCreate projectCreate project.writework.project.create102 connectors
launchdarkly.code_ref_repositories.createCreate repositoryCreate repository.writedevops.repository.create18 connectors

Try it

Prompts to paste into Claude.

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

  1. Readread

    Give me a one-line status for every active LaunchDarkly project.

    callslaunchdarkly.projects.list

  2. Writewrite

    Create team in LaunchDarkly. Show me exactly what you will send before you save it.

    callslaunchdarkly.teams.create

Governance for LaunchDarkly

Destructive actions wait for a yes.

52 LaunchDarkly actions are destructive. The first 10:

  • Delete account memberlaunchdarkly.members.delete
  • Delete teamlaunchdarkly.teams.delete
  • Delete custom rolelaunchdarkly.custom_roles.delete
  • Delete access tokenlaunchdarkly.tokens.delete
  • Delete projectlaunchdarkly.projects.delete
  • Delete repositorylaunchdarkly.code_ref_repositories.delete
  • Delete environmentlaunchdarkly.environments.delete
  • Delete feature flaglaunchdarkly.feature_flags.delete
  • Delete webhooklaunchdarkly.webhooks.delete
  • Delete metriclaunchdarkly.metrics.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.

dolaunchdarkly.members.deleteconfirm_policy: human

Delete account member?

intent: the user asked me to delete account member

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

Cost

What LaunchDarkly costs on actionlake.

Every LaunchDarkly action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call LaunchDarkly 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 401 LaunchDarkly actions
  • 401actions pageeach page of a long list is one call
  • 168reads 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 LaunchDarkly action.

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

Flag approval requests8
  • launchdarkly.flag_approval_requests.applyApply approval request for a flagwrite
  • launchdarkly.flag_approval_requests.createCreate approval request for a flagwrite
  • launchdarkly.flag_approval_requests.create_copyCreate approval request to copy flag configurations across environmentswrite
  • launchdarkly.flag_approval_requests.listList approval requests for a flagread
  • launchdarkly.flag_approval_requests.deleteDelete approval request for a flagdestructive
  • launchdarkly.flag_approval_requests.getGet approval request for a flagread
  • and 2 more in this area
Adaptive triggers7
  • launchdarkly.adaptive_triggers.createCreate an adaptive triggerwrite
  • launchdarkly.adaptive_triggers.disableDisable an adaptive triggerwrite
  • launchdarkly.adaptive_triggers.enableEnable an adaptive triggerwrite
  • launchdarkly.adaptive_triggers.listList adaptive triggersread
  • launchdarkly.adaptive_triggers.deleteDelete an adaptive triggerdestructive
  • launchdarkly.adaptive_triggers.getGet an adaptive triggerread
  • and 1 more in this area
Approval requests7
  • launchdarkly.approval_requests.applyApply approval requestwrite
  • launchdarkly.approval_requests.createCreate approval requestwrite
  • launchdarkly.approval_requests.listList approval requestsread
  • launchdarkly.approval_requests.reviewReview approval requestwrite
  • launchdarkly.approval_requests.deleteDelete approval requestdestructive
  • launchdarkly.approval_requests.getGet approval requestread
  • and 1 more in this area
Environments7
  • launchdarkly.environments.createCreate environmentwrite
  • launchdarkly.environments.getGet environmentread
  • launchdarkly.environments.listList environmentsread
  • launchdarkly.environments.updateUpdate environmentwrite
  • launchdarkly.environments.deleteDelete environmentdestructive
  • launchdarkly.environments.reset_mobile_keyReset environment mobile SDK keywrite
  • and 1 more in this area
Feature flags7
  • launchdarkly.feature_flags.createCreate a feature flagwrite
  • launchdarkly.feature_flags.getGet feature flagread
  • launchdarkly.feature_flags.listList feature flagsread
  • launchdarkly.feature_flags.updateUpdate feature flagwrite
  • launchdarkly.feature_flags.copyCopy feature flagwrite
  • launchdarkly.feature_flags.deleteDelete feature flagdestructive
  • and 1 more in this area
Integration delivery configurations7
  • launchdarkly.integration_delivery_configurations.createCreate delivery configurationwrite
  • launchdarkly.integration_delivery_configurations.listList all delivery configurationsread
  • launchdarkly.integration_delivery_configurations.list_by_environmentGet delivery configurations by environmentread
  • launchdarkly.integration_delivery_configurations.validateValidate delivery configurationwrite
  • launchdarkly.integration_delivery_configurations.deleteDelete delivery configurationdestructive
  • launchdarkly.integration_delivery_configurations.getGet delivery configuration by IDread
  • and 1 more in this area
Flag import configurations6
  • launchdarkly.flag_import_configurations.createCreate a flag import configurationwrite
  • launchdarkly.flag_import_configurations.getGet a single flag import configurationread
  • launchdarkly.flag_import_configurations.listList all flag import configurationsread
  • launchdarkly.flag_import_configurations.triggerTrigger a single flag import runwrite
  • launchdarkly.flag_import_configurations.deleteDelete a flag import configurationdestructive
  • launchdarkly.flag_import_configurations.updateUpdate a flag import configurationwrite
Members6
  • launchdarkly.members.deleteDelete account memberdestructive
  • launchdarkly.members.getGet account memberread
  • launchdarkly.members.listList account membersread
  • launchdarkly.members.updateModify an account memberwrite
  • launchdarkly.members.bulk_updateModify account memberswrite
  • launchdarkly.members.inviteInvite new memberswrite
Relay proxy configs6
  • launchdarkly.relay_proxy_configs.createCreate a new Relay Proxy configwrite
  • launchdarkly.relay_proxy_configs.getGet Relay Proxy configread
  • launchdarkly.relay_proxy_configs.listList Relay Proxy configsread
  • launchdarkly.relay_proxy_configs.resetReset Relay Proxy configuration keywrite
  • launchdarkly.relay_proxy_configs.deleteDelete Relay Proxy config by IDdestructive
  • launchdarkly.relay_proxy_configs.updateUpdate a Relay Proxy configwrite
Release policies6
  • launchdarkly.release_policies.createCreate a release policywrite
  • launchdarkly.release_policies.getGet a release policy by keyread
  • launchdarkly.release_policies.listList release policiesread
  • launchdarkly.release_policies.reorderUpdate the order of existing release policieswrite
  • launchdarkly.release_policies.deleteDelete a release policydestructive
  • launchdarkly.release_policies.replaceUpdate a release policywrite
SDK keys6
  • launchdarkly.sdk_keys.createCreate SDK keywrite
  • launchdarkly.sdk_keys.getGet SDK keyread
  • launchdarkly.sdk_keys.listGet all environment SDK keysread
  • launchdarkly.sdk_keys.list_by_projectGet all project SDK keysread
  • launchdarkly.sdk_keys.deleteDelete SDK keydestructive
  • launchdarkly.sdk_keys.updateUpdate SDK keywrite
Segments6
  • launchdarkly.segments.createCreate segmentwrite
  • launchdarkly.segments.getGet segmentread
  • launchdarkly.segments.listList segmentsread
  • launchdarkly.segments.semantic_patchPatch segment (semantic patch)write
  • launchdarkly.segments.deleteDelete segmentdestructive
  • launchdarkly.segments.updatePatch segmentwrite
Teams6
  • launchdarkly.teams.createCreate teamwrite
  • launchdarkly.teams.getGet teamread
  • launchdarkly.teams.listList teamsread
  • launchdarkly.teams.updateUpdate teamwrite
  • launchdarkly.teams.bulk_updateUpdate teamswrite
  • launchdarkly.teams.deleteDelete teamdestructive
Tokens6
  • launchdarkly.tokens.createCreate access tokenwrite
  • launchdarkly.tokens.deleteDelete access tokendestructive
  • launchdarkly.tokens.getGet access tokenread
  • launchdarkly.tokens.listList access tokensread
  • launchdarkly.tokens.resetReset access token (legacy)write
  • launchdarkly.tokens.updatePatch access tokenwrite
Agent graphs5
  • launchdarkly.agent_graphs.createCreate new agent graphwrite
  • launchdarkly.agent_graphs.getGet agent graphread
  • launchdarkly.agent_graphs.listList agent graphsread
  • launchdarkly.agent_graphs.updateUpdate agent graphwrite
  • launchdarkly.agent_graphs.deleteDelete agent graphdestructive
Agent optimization results5
  • launchdarkly.agent_optimization_results.createCreate agent optimization resultwrite
  • launchdarkly.agent_optimization_results.listList latest agent optimization result per runread
  • launchdarkly.agent_optimization_results.list_allList all agent optimization results across versionsread
  • launchdarkly.agent_optimization_results.list_by_runList agent optimization results for a runread
  • launchdarkly.agent_optimization_results.updateUpdate an agent optimization resultwrite
Agent optimizations5
  • launchdarkly.agent_optimizations.createCreate agent optimizationwrite
  • launchdarkly.agent_optimizations.getGet an agent optimizationread
  • launchdarkly.agent_optimizations.listList agent optimizationsread
  • launchdarkly.agent_optimizations.updateUpdate an agent optimizationwrite
  • launchdarkly.agent_optimizations.deleteDelete an agent optimizationdestructive
Agent skills5
  • launchdarkly.agent_skills.createCreate an agent skillwrite
  • launchdarkly.agent_skills.getGet an agent skillread
  • launchdarkly.agent_skills.listList agent skillsread
  • launchdarkly.agent_skills.updateUpdate an agent skillwrite
  • launchdarkly.agent_skills.deleteDelete an agent skilldestructive
AI configs5
  • launchdarkly.ai_configs.createCreate new AI Configwrite
  • launchdarkly.ai_configs.getGet AI Configread
  • launchdarkly.ai_configs.listList AI Configsread
  • launchdarkly.ai_configs.updateUpdate AI Configwrite
  • launchdarkly.ai_configs.deleteDelete AI Configdestructive
AI tools5
  • launchdarkly.ai_tools.createCreate an AI toolwrite
  • launchdarkly.ai_tools.getGet AI toolread
  • launchdarkly.ai_tools.listList AI toolsread
  • launchdarkly.ai_tools.updateUpdate AI toolwrite
  • launchdarkly.ai_tools.deleteDelete AI tooldestructive
Big segment store integrations5
  • launchdarkly.big_segment_store_integrations.createCreate big segment store integrationwrite
  • launchdarkly.big_segment_store_integrations.getGet big segment store integration by IDread
  • launchdarkly.big_segment_store_integrations.listList all big segment store integrationsread
  • launchdarkly.big_segment_store_integrations.updateUpdate big segment store integrationwrite
  • launchdarkly.big_segment_store_integrations.deleteDelete big segment store integrationdestructive
Code ref repositories5
  • launchdarkly.code_ref_repositories.createCreate repositorywrite
  • launchdarkly.code_ref_repositories.getGet repositoryread
  • launchdarkly.code_ref_repositories.listList repositoriesread
  • launchdarkly.code_ref_repositories.updateUpdate repositorywrite
  • launchdarkly.code_ref_repositories.deleteDelete repositorydestructive
Custom roles5
  • launchdarkly.custom_roles.createCreate custom rolewrite
  • launchdarkly.custom_roles.getGet custom roleread
  • launchdarkly.custom_roles.listList custom rolesread
  • launchdarkly.custom_roles.updateUpdate custom rolewrite
  • launchdarkly.custom_roles.deleteDelete custom roledestructive
Destinations5
  • launchdarkly.destinations.createCreate Data Export destinationwrite
  • launchdarkly.destinations.getGet destinationread
  • launchdarkly.destinations.listList destinationsread
  • launchdarkly.destinations.updateUpdate Data Export destinationwrite
  • launchdarkly.destinations.deleteDelete Data Export destinationdestructive
Experiments5
  • launchdarkly.experiments.createCreate experimentwrite
  • launchdarkly.experiments.getGet experimentread
  • launchdarkly.experiments.listGet experimentsread
  • launchdarkly.experiments.updatePatch experimentwrite
  • launchdarkly.experiments.list_any_environmentGet experiments any environmentread
Flag scheduled changes5
  • launchdarkly.flag_scheduled_changes.createCreate scheduled changes workflowwrite
  • launchdarkly.flag_scheduled_changes.getGet a scheduled changeread
  • launchdarkly.flag_scheduled_changes.listList scheduled changesread
  • launchdarkly.flag_scheduled_changes.updateUpdate scheduled changes workflowwrite
  • launchdarkly.flag_scheduled_changes.deleteDelete scheduled changes workflowdestructive
Flag triggers5
  • launchdarkly.flag_triggers.createCreate flag triggerwrite
  • launchdarkly.flag_triggers.getGet flag trigger by IDread
  • launchdarkly.flag_triggers.listList flag triggersread
  • launchdarkly.flag_triggers.updateUpdate flag triggerwrite
  • launchdarkly.flag_triggers.deleteDelete flag triggerdestructive
Holdouts5
  • launchdarkly.holdouts.createCreate holdoutwrite
  • launchdarkly.holdouts.getGet holdoutread
  • launchdarkly.holdouts.get_by_idGet Holdout by Idread
  • launchdarkly.holdouts.listGet all holdoutsread
  • launchdarkly.holdouts.updatePatch holdoutwrite
Insights charts5
  • launchdarkly.insights_charts.deployment_frequencyGet deployment frequency chart dataread
  • launchdarkly.insights_charts.flag_statusGet flag status chart dataread
  • launchdarkly.insights_charts.lead_timeGet lead time chart dataread
  • launchdarkly.insights_charts.release_frequencyGet release frequency chart dataread
  • launchdarkly.insights_charts.stale_flagsGet stale flags chart dataread
Insights groups5
  • launchdarkly.insights_groups.createCreate insight groupwrite
  • launchdarkly.insights_groups.getGet insight groupread
  • launchdarkly.insights_groups.listList insight groupsread
  • launchdarkly.insights_groups.updatePatch insight groupwrite
  • launchdarkly.insights_groups.deleteDelete insight groupdestructive

Integration configurations 5Integration subscriptions 5Metric groups 5Metrics 5Model configs 5Oauth clients 5Projects 5Prompt snippets 5Release pipelines 5Views 5Warehouse destinations 5Webhooks 5AI config variations 4Announcements 4Applications 4Code ref branches 4Flag followers 4Flag links 4Flag releases 4Flag workflows 4Application versions 3Audit log entries 3Feature flag statuses 3Insights deployments 3Ip allowlist entries 3Layers 3Project flag defaults 3User flag settings 3Users 3View linked resources 3Workflow templates 3Agent optimization runs 2AI config metrics 2AI config targeting 2AI restricted models 2Approval request settings 2Big segment exports 2Code ref extinctions 2Code ref statistics 2Context attributes 2Context kinds 2Contexts 2Dependent flags 2Experimentation settings 2Flag expiring context targets 2Flag expiring user targets 2Insights repository projects 2Ip allowlist 2Segment expiring context targets 2Segment expiring user targets 2User expiring flags 2Agent skill references 1Agent skill versions 1AI config quick stats 1AI runs usage 1AI tool references 1AI tool versions 1API versions 1Audit log entry counts 1Big segment context targets 1Big segment imports 1Big segment user targets 1Caller identity 1Context flag settings 1Context instance segment memberships 1Context instances 1Contexts clientside usage 1Contexts serverside usage 1Contexts total usage 1Data export events usage 1Data export project rankings 1Evaluations usage 1Events usage 1Experiment iterations 1Experimentation events usage 1Experimentation keys usage 1Flag release phases 1Insights deployment events 1Insights flag events 1Insights pull requests 1Insights repositories 1Insights scores 1Linked views 1Mau clientside usage 1Mau SDKS by types 1Mau total usage 1Mau usage 1Mau usage by categories 1Member teams 1Migration safety issues 1Model config versions 1Observability errors usage 1Observability logs usage 1Observability metrics usage 1Observability sessions usage 1Observability traces usage 1Openapi spec 1Prompt snippet references 1Prompt snippet versions 1Public ip list 1Release pipeline progressions 1Root 1SDK all versions 1SDK key views 1SDK versions details 1Segment context memberships 1Segment user memberships 1Service connections usage 1Stream usage 1Stream usage by SDK versions 1Stream usage sdkversions 1Tags 1Team maintainers 1Team roles 1Vega AI usage 1Warehouse export usage 1

This index shows a sample of each area. All 401 actions, with their inputs and prices, are in the catalogue, which a free sandbox key can search.

Questions

The LaunchDarkly MCP server, answered.

Is there a LaunchDarkly MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 401 LaunchDarkly actions across flag approval requests, adaptive triggers, approval requests and environments, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by LaunchDarkly.

What do I need to connect LaunchDarkly?

LaunchDarkly API access token (personal or service token). Plus your host. The free call account { op: "connections.requirements", service: "launchdarkly" } lists exactly what to paste.

Can an AI agent delete or change things in LaunchDarkly?

Only if its key allows it, and only with an explicit confirmation. 52 LaunchDarkly actions are destructive, including delete account member, delete team and delete custom role. 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 LaunchDarkly API limit?

actionlake paces every call to the limits in the LaunchDarkly service file: 10 calls a second per connection. Repeated reads are answered from cache (168 LaunchDarkly actions can be), which never reaches LaunchDarkly.

What does a LaunchDarkly action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (401 LaunchDarkly actions page). Cache hits, searches, dry runs and calls LaunchDarkly 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 LaunchDarkly API does it cover?

The LaunchDarkly REST API v2 (OpenAPI 3.0.3) at app.launchdarkly.com by default, or the base URL set when you connect. It is built from LaunchDarkly's documentation as read on 22 October 2026. 401 actions in 146 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my LaunchDarkly 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 LaunchDarkly 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 LaunchDarkly today.

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