How is this different from connecting LaunchDarkly MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the LaunchDarkly MCP API changes?

We track the LaunchDarkly MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different LaunchDarkly tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in LaunchDarkly?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the LaunchDarkly MCP server myself?

No. TrustGate connects to the LaunchDarkly MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to LaunchDarkly directly.

Connect LaunchDarkly MCP to your AI agents, securely.

Q: Is LaunchDarkly MCP secure?

The server itself exposes 170 tools, and some of them can write to your LaunchDarkly account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches LaunchDarkly.

One secured endpoint for every LaunchDarkly call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try LaunchDarkly MCP now

What your agents can ask once LaunchDarkly MCP is connected.

Select the tools your AI agents can call

170 tools from the LaunchDarkly MCP schema.

Showing 170 tools

delete-flaghigh riskDeletes a feature flag from a project.
archive-flaghigh riskArchive a feature flag (reversible). This is a soft-delete that can be undone.
copy-flag-confighigh riskCopy a flag's targeting configuration from one environment to another.
toggle-flaghigh riskTurns a flag's targeting on or off in a specific environment. This takes effect immediately.
update-flag-settingswriteUpdate a flag's top-level settings: name, description, tags, temporary/permanent status, maintainer, default variations, or custom properties.
create-flagwriteCreates a new feature flag in a project.
get-flagreadGets the detailed configuration of a single flag, scoped to a specific environment.
list-flagsreadSearches and browses the feature flags in a project, with filtering.
match-release-policiesreadResolve which release policies would govern a flag in a given environment — a read-only dry-run of the server-side policy matching logic.
create-automated-rollout-configwriteRecord an automated rollout config in LaunchDarkly for a feature-flagged change.
stop-guarded-rollouthigh riskStop an active guarded rollout on a flag's default rule (fallthrough).
start-guarded-rolloutwriteStart a guarded rollout on a flag's default rule (fallthrough).
list-release-policiesreadList release policies for a project.
manage-expiring-targetshigh riskList, add, update, or remove expiring targets on a flag.
update-prerequisiteshigh riskAdd or remove prerequisites for a flag in a specific environment.
update-individual-targetshigh riskAdd or remove specific users or contexts from individual flag targeting.
update-targeting-ruleshigh riskAdd, remove, or modify custom targeting rules for a flag in an environment.
update-off-variationhigh riskSet the variation a flag serves when its targeting is OFF, in one environment (env), several (envs), or every environment (allEnvs).
update-rollouthigh riskChange the default rule (fallthrough) for a flag.
get-flag-unique-context-exposuresreadGet unique-context exposure counts (distinct-context cardinality) for a flag in an environment.
check-flag-rollout-completereadCheck whether a feature flag is fully rolled out to a single deterministic variation across all traffic-bearing environments.
query-flag-evaluationsreadQuery flag evaluations for a session. Retrieve flag evaluation events for a specific session to understand which feature flags were evaluated during that session.
get-flag-status-across-envsreadChecks a flag's lifecycle status across all environments in a project.
get-flag-healthreadGet combined health view for a single flag: lifecycle state, last-requested timestamp, targeting summary (rules count, targets count, on/off), age, and whether it's temporary.
check-removal-readinessreadRuns a safety check before removing a flag: dependencies, code references, active targeting and expiring rules.
find-stale-flagsreadFinds flags that are candidates for cleanup, sorted by staleness.
update-segment-targetshigh riskAdd or remove individual context keys from a segment's included or excluded lists.
update-segment-ruleshigh riskAdd, remove, or modify attribute-based targeting rules on a segment.
get-segmentreadGets a segment with its rules and membership.
create-segmentwriteCreates a reusable segment of contexts that flags can target.
list-segmentsreadLists the segments in a project and environment.
list-projectsreadLists the projects in the LaunchDarkly account.
get-sdk-activereadCheck whether any SDKs have been active in a given environment.
get-environmentreadGet a LaunchDarkly environment's client-side ID and mobile key.
get-projectreadGets a project with its environments and SDK keys.
create-projectwriteCreate a new LaunchDarkly project. Returns the project with its environments, including each environment's client-side ID and mobile key.
get-experiment-metric-resultsreadGet detailed statistical results for a single metric on an experiment iteration.
get-experiment-resultsreadGet an overall results summary for an experiment, across every metric on the current iteration.
stop-experiment-iterationhigh riskStop the currently running iteration of an experiment.
save-and-start-experiment-iterationhigh riskStop the current running iteration, create a new draft iteration with the provided field updates applied, and start it.
start-experiment-iterationhigh riskStart an experiment iteration. This begins data collection for the experiment's current draft iteration and starts allocating live end-user traffic across its treatments.
update-experimenthigh riskUpdate fields on an experiment or its current iteration.
create-experimentwriteCreate a new experiment on a flag or AgentControl Config.
get-experimentreadGet detailed information about a specific experiment, including its treatments, metrics, and current iteration status.
list-experimentsreadList experiments in a project, optionally filtered by environment.
list-metric-eventsreadList recent event keys received by a LaunchDarkly project — used to check which events are actively flowing before creating a metric.
create-metricwriteCreate a new metric in a LaunchDarkly project.
get-metricreadGet full details for a single LaunchDarkly metric: measureType, eventKey, successCriteria, unit, description, tags, creation date, and attached flag count.
list-metricsreadList metrics in a LaunchDarkly project. Returns key, name, measureType (count/occurrence/value), eventKey, successCriteria, tags, and how many flags each metric is attached to.
query-change-historyreadQuery the audit log to find what changed in a LaunchDarkly account.
apply-approval-requesthigh riskApply an already-approved flag or segment approval request.
list-approval-requestsreadList approval requests for a flag or a segment in an environment.
create-approval-requestwriteCreate an approval request for a flag or segment change in an environment that requires approvals.
unrestrict-agentcontrol-model-configshigh riskRemove AI model configs from the restricted list so that variations can select them again.
restrict-agentcontrol-model-configshigh riskRestrict AI model configs so that variations cannot select them.
delete-agentcontrol-model-confighigh riskPermanently delete a custom AI model config.
update-agentcontrol-model-configwriteUpdate a custom AI model config. modelConfigKey is the key from list-agentcontrol-model-configs (for example OpenAI.gpt-image-2), not the id.
create-agentcontrol-model-configwriteCreate a custom AI model config for a model that is not in the global catalog.
get-agentcontrol-model-configreadGet one AI model config. modelConfigKey is the key from list-agentcontrol-model-configs (for example OpenAI.gpt-4o), not the id.
list-agentcontrol-model-configsreadList the AI model configs in a project: global catalog models (global=true) and custom models (global=false).
list-agent-skill-referencesreadList the AgentControl Config variations that attach an agent skill, with the skill version each one is pinned to.
list-agent-skill-versionsreadList the versions of an agent skill, newest first. Returns each version's number, name, description, tags, and creation time.
delete-agent-skillhigh riskPermanently delete an agent skill and all of its versions.
update-agent-skillwriteUpdate an agent skill's markdown, name, description, tags, or maintainer.
create-agent-skillwriteCreate an agent skill from SKILL.md content.
get-agent-skillreadGet the latest version of an agent skill by key, including its full SKILL.md markdown, name, description, tags, and version number.
list-agent-skillsreadList agent skills in a project.
list-ai-tool-versionsreadList the versions of one AI tool in a project, newest first.
delete-prompt-snippethigh riskPermanently delete a prompt snippet.
update-prompt-snippetwriteUpdate an existing prompt snippet. Creates a new version of the snippet.
create-prompt-snippetwriteCreate a new prompt snippet. A snippet is a reusable text block that can be referenced in AgentControl Config variation prompts using {{snippet-key}} syntax.
get-prompt-snippetreadGet a specific prompt snippet by key. Returns the snippet's name, text content, tags, version, and creation time.
list-prompt-snippetsreadList prompt snippets in a project. Prompt snippets are reusable text blocks that can be referenced inside AgentControl Config variation prompts to keep common instructions consistent.
delete-ai-toolhigh riskPermanently delete an AI tool definition.
update-ai-toolwriteUpdate an AI tool definition's description, schema, tags, or customParameters.
toggle-agentcontrol-confighigh riskNot supported: always fails with 400. AgentControl Config targeting cannot be toggled; the API rejects turnFlagOn/turnFlagOff.
preview-graphreadPreview a chart/graph inline WITHOUT saving it to a dashboard.
create-graphwriteAdd a chart/graph to an existing dashboard.
delete-agent-graphhigh riskPermanently delete an agent graph and all of its edges.
update-agent-graphhigh riskUpdate an agent graph's metadata or structure.
create-agent-graphwriteCreate a new agent graph in a project. An agent graph defines a directed graph of AgentControl Configs for multi-agent workflows.
get-agent-graphreadGet a specific agent graph by key, including its full edge structure.
list-agent-graphsreadList all agent graphs in a project. Returns metadata for each graph including key, name, description, and root config key.
create-ai-toolwriteCreate a new AI tool definition in a project.
get-ai-toolreadGet a single AI tool definition including its full schema, version, tags, and customParameters.
list-ai-toolsreadList AI tool definitions in a project. Returns each tool's key, description, schema, version, tags, and customParameters.
update-agentcontrol-config-targetshigh riskAdd or remove specific contexts in AgentControl Config individual targeting in one environment.
update-agentcontrol-config-rollouthigh riskSet the default (fallthrough) rule for an AgentControl Config in one environment.
update-agentcontrol-config-targetinghigh riskChange custom targeting rules for an AgentControl Config in one environment.
get-agentcontrol-config-targetingreadRead the targeting configuration for an AgentControl Config in a specific environment.
clone-agentcontrol-config-variationwriteCopy an existing variation. Use update-agentcontrol-config-variation to change it in place.
delete-agentcontrol-config-variationhigh riskPermanently delete an AgentControl Config variation.
update-agentcontrol-config-variationhigh riskUpdate variation. Requires projectKey, configKey, variationKey, and a change.
create-agentcontrol-config-variationwriteCreate a variation in an existing config from create-agentcontrol-config.
get-agentcontrol-config-healthreadHealth check for an AgentControl Config.
get-agentcontrol-config-status-envsreadCheck an AgentControl Config's lifecycle status across all environments.
setup-agentcontrol-configwriteCreate an AgentControl Config and its first variation in one step, then verify it.
delete-agentcontrol-confighigh riskPermanently delete an AgentControl Config.
update-agentcontrol-configwriteUpdate an AgentControl Config's name, description, tags, default variation, judge metric, judge success direction (isInverted), or maintainer.
create-agentcontrol-configwriteCreate an AgentControl Config shell in a project.
get-agentcontrol-configreadRead config. Requires projectKey, configKey. Find configKey with list-agentcontrol-configs.
list-agentcontrol-configsreadSearch AgentControl Configs in a project.
apply-agentcontrol-config-approvalhigh riskApply an already-approved AgentControl Config approval request.
list-agentcontrol-config-approvalsreadList approval requests for an AgentControl Config.
create-agentcontrol-config-approvalwriteCreate an approval request for an AgentControl Config change.
update-playgroundwriteUpdate an LLM playground's name, variants, or archived status.
create-playgroundwriteCreate a new LLM playground. A playground lets you compare AgentControl Config variations side-by-side.
get-playgroundreadGet details about a specific LLM playground, including its variants (evaluation definitions and their positions).
list-playgroundsreadList LLM playgrounds in a project. Returns playground names, variant counts, and timestamps.
get-evaluation-run-summaryreadGet the progress and results of an evaluation run.
run-evaluationwriteStart a run of an AI evaluation. evaluationId comes from create-evaluation or list-evaluations, and the evaluation must use only deepeval criteria.
create-evaluationwriteCreate an AI evaluation: a saved prompt, a generation model, and scoring criteria.
get-evaluationreadGet one AI evaluation by evaluationId (from list-evaluations or create-evaluation).
list-evaluationsreadList AI evaluations in a project with id, version, name, models, messages, criteria, and totalCount.
delete-datasethigh riskPermanently delete an offline dataset and its associated metadata.
create-datasetwriteCreate an offline dataset for AI evaluation.
get-datasetreadGet one offline dataset by id with key, name, filename, format, status, rowCount, and createdAt.
list-datasetsreadList offline datasets in a project with id, key, name, filename, format, status (pending, ready, or failed), rowCount, and createdAt.
get-session-timestamp-event-chunk-urlreadGet a signed download URL for the rrweb event chunk covering a specific timestamp within a session.
get-session-event-chunk-urlreadGet a signed download URL for one rrweb event chunk of a session.
get-session-event-chunksreadList the rrweb event chunks recorded for a session.
query-service-mapreadQuery the service dependency map (service map) for a project.
get-service-connections-usagereadGet a time series of service connection minutes from your LaunchDarkly account.
query-timeline-eventsreadQuery timeline indicator events for a session.
query-sessionsreadQuery project sessions. Retrieve sessions for a given project with explicit date range parameters.
query-error-groupsreadQuery project error groups. Retrieve error groups for a given project with explicit date range parameters.
query-aggregationsreadRetrieve bucketed, aggregated values over time for a product type.
query-tracesreadQuery project traces. Retrieve traces for a given project with explicit date range parameters.
query-logsreadQuery project logs. Retrieve logs for a given project with explicit date range parameters.
get-resource-viewsreadList the views a flag, segment, or AgentControl Config is linked to.
list-view-resourcesreadList the resources of one type linked to a view.
link-view-resourceswriteLink flags, segments, or AgentControl Configs to a view.
update-viewwriteUpdate a view's name, description, maintainer, or tags.
create-viewwriteCreate a new view in a project. Views are an Enterprise feature.
get-viewreadGet a single view by key. Views are an Enterprise feature. Returns name, description, tags, maintainer, and linked-resource counts (flags, segments, AgentControl Configs).
list-viewsreadList views in a project. Views are an Enterprise feature. Returns a paginated list with key, name, description, tags, maintainer, and (when includeResourceSummary is set) linked-resource counts.
delete-alerthigh riskPermanently delete an observability alert.
toggle-alerthigh riskEnable or disable an observability alert without changing its configuration.
update-alerthigh riskUpdate an existing observability alert. Only the fields you provide are changed; everything else is preserved.
create-alertwriteCreate a new observability alert that fires when a metric crosses a threshold.
get-alert-firing-historyreadShow when an alert actually fired, and what state it is in now.
get-alertreadGet detailed information about a specific observability alert.
list-alertsreadList existing observability alerts for the project.
delete-dashboardhigh riskPermanently delete a dashboard (visualization) and all of its graphs.
get-keysreadDiscover available data keys/dimensions for a product type.
create-dashboardwriteCreate a new empty dashboard (visualization) for organizing charts.
get-dashboardreadGet detailed information about a specific dashboard including all its graphs.
list-dashboardsreadList existing dashboards (visualizations) for the project.
delete-factory-repo-settingshigh riskDeprecated: use delete-release-autopilot-repo-settings instead.
update-factory-repo-settingshigh riskDeprecated: use update-release-autopilot-repo-settings instead.
get-factory-repo-settingsreadDeprecated: use get-release-autopilot-repo-settings instead.
list-factory-github-reposreadDeprecated: use list-release-autopilot-github-repos instead.
list-factory-repo-settingsreadDeprecated: use list-release-autopilot-repo-settings instead.
update-factory-settingshigh riskDeprecated: use update-release-autopilot-settings instead.
get-factory-settingsreadDeprecated: use get-release-autopilot-settings instead.
delete-release-autopilot-repo-settingshigh riskUnmap a GitHub repository from Release Autopilot settings (removes project mapping and overrides).
update-release-autopilot-repo-settingshigh riskMap a GitHub repository to a LaunchDarkly project and optionally override auto-flagging or auto-releasing.
get-release-autopilot-repo-settingsreadGet Release Autopilot settings for one GitHub repository (effective auto-flagging / auto-releasing).
list-release-autopilot-github-reposreadList GitHub repositories on this account's LaunchDarkly GitHub App install, with display name (usually owner/name) and numeric githubRepoId.
list-release-autopilot-repo-settingsreadList Release Autopilot settings for GitHub repositories already mapped to a LaunchDarkly project.
update-release-autopilot-settingshigh riskUpdate account-wide Release Autopilot settings (auto-flagging and auto-releasing).
get-release-autopilot-settingsreadGet account-wide Release Autopilot settings: whether auto-flagging and auto-releasing are on, and whether auto-flagging pull requests need approval.
ventwriteReport a missing capability, bug, parameter gap, or unclear error encountered while using LaunchDarkly MCP tools.
get-docreadFetch the full markdown content of a LaunchDarkly documentation page.
search-docsreadSearch LaunchDarkly documentation to retrieve authoritative reference material for flags, targeting rules, experiments, segments, AgentControl configs, SDKs, API fields, rollouts, metrics, and more.
get-context-instancesreadGet all stored instances of a specific context by kind and key.
search-contextsreadSearch for stored LaunchDarkly contexts in an environment by kind, key, or attribute value.
invite-membershigh riskInvite one or more people to the LaunchDarkly account by email.
find-membersreadSearch for LaunchDarkly account members and return their IDs.
get-member-selfreadReturn the member ID, profile, and team memberships for the currently authenticated user.

What LaunchDarkly MCP exposes

LaunchDarkly MCP groups its 170 tools into 14 categories, each with a suggested starting policy.

All 170 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Flags	Flag listing, details, creation, settings, toggling, copying configuration, archiving and deletion	Read / Write	Require approval for every flag write; a toggle changes production behaviour instantly
Targeting & Rollouts	Targeting rules, individual targets, prerequisites, off variations, percentage, automated and guarded rollouts, expiring targets and release policies	Read / Write	Require approval; targeting changes alter production behaviour instantly
Flag Status	Flag status across environments, flag health, rollout completion, context exposures and flag evaluation data	Read	Allow read-only status checks
Lifecycle	Stale flag detection and removal readiness checks	Read	Run the readiness check before any removal
Segments	Segment listing, details, creation, rule updates and target updates	Read / Write	Segment changes shift who sees a feature; audit them
Projects & Environments	Project listing, details and creation, environments and SDK activity	Read / Write	Project and environment reads can return SDK keys; treat them as credentials
Experiments & Metrics	Experiments and their iterations and results, and metrics and metric events	Read / Write	Approve starting or stopping iterations; results feed release decisions
Approvals & History	Flag approval requests, applying approved changes and change history	Read / Write	Keep applying approvals with a human
AgentControl	AgentControl configs, variations, targeting, rollouts and approvals, model configs, prompt snippets, AI tools, agent graphs and agent skills	Read / Write	Require approval for config, targeting and model changes; they change live AI behaviour
AI Evaluations	Playgrounds, evaluations and their runs, and evaluation datasets	Read / Write	Allow in non-production projects; datasets can hold sample user data
Observability	Querying logs, traces, error groups, sessions, timeline events, aggregations, the service map and service connection usage	Read	Logs and sessions can contain personal data; limit to investigating agents
Alerts & Dashboards	Observability alerts and their firing history, dashboards, data keys and views	Read / Write	Audit alert changes; a muted alert hides incidents
Release Autopilot	Release Autopilot account settings and GitHub repository mappings, including the deprecated factory tools	Read / Write	Approve settings changes; they control automatic flagging and releasing
Account, Contexts & Docs	Member lookup and invitations, contexts and their instances, documentation search and tool feedback	Read / Write	Require approval for invitations; they grant account access

Connect LaunchDarkly MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect LaunchDarkly
Find LaunchDarkly in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow LaunchDarkly MCP or block it. You allow the exact tools each agent can call, one by one.

LaunchDarkly MCP	170 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 170 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is LaunchDarkly MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	LaunchDarkly MCP170 documented tools.
Coding assistants	Connection & credentialsper user	AgentControl52 tools
Internal copilots	Tool access policiesdeny by default	Alerts & Dashboards19 tools
Workflow automations	Request & response inspectioninline	Release Autopilot14 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools85 tools