Unleash Mcp
Connects your LLM coding assistant directly to the Unleash Admin API for creating and managing feature flags without leaving your workflow. You get tools to evaluate whether a code change needs flagging, detect existing flags to avoid duplicates, create new flags with proper validation, and generate language-specific wrapper code. It handles the full lifecycle: checking rollout strategies, toggling flags across environments, and eventually cleaning up flag code when you're ready to remove it. Built around Unleash's best practices for flag management at scale, so your assistant knows when to recommend gradual rollouts versus instant deploys. Works with stdio transport locally or connects to Unleash instances running the experimental remote MCP server over HTTP.
- GitHub stars
- 21
- tools
- 11
- auth
- Secret needed
- transport
- STDIO
- env vars
- 5
- last updated
- Nov 6, 2026
Tools
11 toolsPublic tool metadata for what this MCP can expose to an agent.
searchFind projects, flags, segments, context fields, tags, tag types, flag creators, release templates, event types and environments by name.
Find projects, flags, segments, context fields, tags, tag types, flag creators, release templates, event types and environments by name.
No parameter schema in public metadata yet.
inspectRead a project, flag, context field or segment in full, or a flag's exposure metrics in an environment.
Read a project, flag, context field or segment in full, or a flag's exposure metrics in an environment.
No parameter schema in public metadata yet.
evaluateTell whether a flag is enabled, disabled or unknown for a context in an environment.
Tell whether a flag is enabled, disabled or unknown for a context in an environment.
No parameter schema in public metadata yet.
releaseTurn a flag on for audiences, split by value or through a release template, or turn it off. Creates the flag if it doesn't exist. Opens a change request where the environment requires one.
Turn a flag on for audiences, split by value or through a release template, or turn it off. Creates the flag if it doesn't exist. Opens a change request where the environment requires one.
No parameter schema in public metadata yet.
kill_switchTurn a permanent kill switch on or off, for everyone or chosen audiences. Creates the switch if it doesn't exist. Opens a change request where the environment requires one.
Turn a permanent kill switch on or off, for everyone or chosen audiences. Creates the switch if it doesn't exist. Opens a change request where the environment requires one.
No parameter schema in public metadata yet.
experimentRun an A/B test: give audiences values such as control and treatment, by share or one each, or stop it. Creates a missing flag. Opens a change request where the environment requires one.
Run an A/B test: give audiences values such as control and treatment, by share or one each, or stop it. Creates a missing flag. Opens a change request where the environment requires one.
No parameter schema in public metadata yet.
manage_flagDefine flags, set a parent dependency, mark flags completed, archive and revive them. Doesn't delete flags.
Define flags, set a parent dependency, mark flags completed, archive and revive them. Doesn't delete flags.
No parameter schema in public metadata yet.
configureChange context fields and their legal values, segments, tag types and tags, release templates and projects.
Change context fields and their legal values, segments, tag types and tags, release templates and projects.
No parameter schema in public metadata yet.
auditRead what changed, who changed it and when, over at most 31 days.
Read what changed, who changed it and when, over at most 31 days.
No parameter schema in public metadata yet.
docsAnswer questions about feature flag practices and Unleash itself from the current Unleash documentation.
Answer questions about feature flag practices and Unleash itself from the current Unleash documentation.
No parameter schema in public metadata yet.
send_feedbackReport to Unleash where the server fell short. Asks you first, unless `UNLEASH_MCP_SEND_FEEDBACK` decides.
Report to Unleash where the server fell short. Asks you first, unless `UNLEASH_MCP_SEND_FEEDBACK` decides.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
UNLEASH_BASE_URL*Your Unleash instance URL (local or hosted)
UNLEASH_PAT*secretYour Unleash Personal Access Token (PAT) for authentication
UNLEASH_MCP_SEND_FEEDBACKtrue or false to grant or deny send_feedback consent without a prompt
UNLEASH_FEEDBACK_URLUnleash instance that receives send_feedback reports. Unset, Unleash's instance (https://eu.app.unleash-hosted.com/hosted/).
UNLEASH_MCP_CONFIG_DIRDirectory holding feedback_consent.json (defaults to the platform config directory)
Readme
Unleash MCP server
An MCP server that lets AI agents work with Unleash the way you'd work in the Unleash UI: find flags, roll them out, check who gets them, and see what changed. It's in alpha and isn't published to npm, so you run it from source.
Tools
Details
- Repository
- github.com/unleash/unleash-mcp
- Category
- AI & LLM Tools
- Package
@unleash/mcp- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Codex Mcp Toolx51xxx/codex-mcp-toolMCP server bridging AI assistants to OpenAI Codex CLI for code analysis and review21stars
hmem — Humanlike Memory for AI Agentsbumblebiber/hmemPersistent 5-level hierarchical memory for AI agents. SQLite-backed, lazy-loaded.20stars
Kirbah Mcp Youtubekirbah/mcp-youtubeProvide token-optimized, structured YouTube data to enhance your LLM applications. Access efficien…19stars
NeuroDock Cognitive Graphtlennon-ie/neurodockA local typed-edge graph of people, projects, and decisions that externalises memory.18stars
Mcp As A Judgeothervibes/mcp-as-a-judgeMCP as a Judge: a behavioral MCP that strengthens AI coding assistants via explicit LLM evaluations17stars
ToolTrust Scanneragentsafe-ai/tooltrust-scannerScans MCP servers for prompt injection, data exfiltration, and privilege escalation.16stars