Now in public beta

Code moves at AI speed. Your control layer should too.

Feature flags, remote config, and AI prompt management that ship as fast as your team.

Free forever for small projects. No credit card required.

flagdash.io/dashboard
LIVE
Feature Flags production
new-checkout
dark-mode click me
beta-search
ai-copilot 25%
API Response 2ms
GET /api/v1/flags
{ "flags": [ { "key": "new-checkout", "value": true }, { "key": "dark-mode", "value": false }, { "key": "beta-search", "value": true }, { "key": "ai-copilot", "value": true } ] }
<3ms

API Response Time

99.9%

Uptime SLA

3-Tier

API Security

0s

Deploy Time

MCP Session
27 tools
"Enable dark-mode for 25% of users in production"
ToggleFlag
dark-mode: enabled
SetFlagRollout
rollout: 25%
Done. dark-mode is now enabled for 25% of production users. The rollout uses consistent hashing — the same users will stay in the cohort as you increase the percentage.
Works with
Claude Code Cursor Windsurf ChatGPT
MCP Server

Your AI writes the flags.
You just describe what you want.

FlagDash speaks MCP natively. Connect Claude, Cursor, or ChatGPT and manage your entire feature flag infrastructure through natural conversation. 27 tools, zero context switching.

  • Create, toggle, and rollout flags from your IDE
  • One-line setup — just claude mcp add
  • Flags, configs & AI configs — all via conversation
AI Context Loading

Say "use ai configs"
and your AI knows everything.

Create agent personas, skills, and rules in the dashboard. When you need them, just tell the AI to load your configs. Your instructions are injected directly into the LLM context — no code, no setup, no prompts to copy-paste.

Update behavior from the dashboard — no redeploy
Different configs per environment (staging vs production)
Works with Claude Code, Cursor, Windsurf, ChatGPT
"Use ai configs, then help me write tests for the auth module"
UseAiConfigs
loaded: agent.md, code-review.md, tone.md
Context injected
agent.md — "You are a senior Elixir engineer…"
code-review.md — skill
tone.md — rule
Got it. I’ve loaded your project context. Let me review the auth module and write ExUnit tests following your code review guidelines…
Feature Flags

Ship code behind flags.
Release when ready.

Boolean, string, number, and JSON flags with instant toggling. Deploy code behind flags and control releases without redeploying.

  • Instant toggle — no deploy needed
  • Percentage-based rollouts
  • Boolean, string, number & JSON types
Feature Flags production
new-dashboard boolean
checkout-v2 string
legacy-api boolean
pricing-test 50% rollout
Remote Config
editing... staging
{ "pricing": { "free_tier_limit": 10000, "premium_price": 5.99, "trial_days": 14, "currency": "USD" }, "onboarding": { "show_tour": true, "welcome_msg": "Welcome!" } }
Remote Config

Update settings on the fly.
No code changes required.

JSON-powered configuration that updates instantly. Change pricing, copy, feature limits, and more — all from the dashboard.

  • Structured JSON with validation
  • Environment-specific overrides
  • Instant propagation via API
AI Config

Ship AI prompts like code.
Version per environment.

Manage agent instructions, skills, and rules as markdown files — each environment gets its own set. Update your AI's behavior instantly without redeploying.

  • Agent, skill, and rule file types
  • MCP server for direct AI agent access
  • Folder organization & markdown editor
AI Config 3 files
production
All Types Agent Skill Rule

code-review.md

skill
# Code Review Rules Review all pull requests for security...

safety-guidelines.md

rule
# Safety Guidelines Never execute destructive operations...
agents (1 file)

agent.md

agent
# Agent Instructions You are an AI assistant working within this...
Available via MCP Server — AI agents read configs automatically
SSE Stream
connected
flag_updated 12:04:32.001
new-checkout: true → all clients refreshed
config_updated 12:04:33.450
pricing-tiers: updated → UI refreshed
ai_config_updated 12:04:35.220
agent.md: modified → agents notified
ping 12:04:55.000
Realtime

Changes propagate instantly.
Zero polling required.

Toggle a flag in the dashboard and every connected client updates in milliseconds via Server-Sent Events. No refresh, no delay, no stale state.

  • SSE push — flags, configs & AI configs
  • Auto-reconnect with exponential backoff
  • Polling fallback for server SDKs

SDKs for every stack

First-class clients for JavaScript, React, Node.js, Go, and Python. Consistent API surface across all languages.

JS

JavaScript

@flagdash/sdk

SSE Polling

React

@flagdash/react

7 Hooks SSE
N

Node.js

@flagdash/node

Server Mgmt
Go

Go

flagdash-go

Zero deps Mgmt
Py

Python

flagdash

Async Mgmt
3 lines to get started React
import { useFlag } from '@flagdash/react' ;
const enabled = useFlag ( 'new-checkout' , false ) ;
return enabled ? <NewCheckout /> : <OldCheckout /> ;

Everything you need to ship safely

From API security to audit logs, every tool to manage features at scale.

Multi-Environment

Separate flag states for development, staging, and production. Test safely before rolling out.

3-Tier API Security

Client, Server, and Management keys with distinct scopes. Client keys are safe for frontend.

Targeting & Rollouts

Target users by attributes, segments, and custom rules. Percentage-based deployments.

Webhooks

HMAC-signed webhooks for every flag change. Integrate with your deployment pipeline.

Audit Logs

Full trail of who changed what, and when. Track every flag toggle and config update.

API-First

RESTful API with sub-3ms responses. Integrate with any language. Management API for automation.

A dashboard you'll actually enjoy

Manage all your flags, configs, and environments from a clean, intuitive interface.

flagdash.io/dashboard/flags

Feature Flags

+ New Flag
new-dashboard boolean
checkout-v2 string
legacy-api boolean
pricing-experiment 50% rollout
New — CLI / TUI

Manage everything from your terminal

A native Rust TUI built with Ratatui. Create flags, toggle rollouts, update configs, and manage AI prompts — all without leaving your keyboard.

  • Full CRUD for flags, configs, AI configs & webhooks
  • Vi-style navigation — j/k, /, Enter, Esc
  • macOS · Linux · Windows — amd64 & arm64
  • Homebrew, npm, Cargo, or shell script install
flagdash
[2] Flags
[3] Config
[4] AI Config
new-checkout bool true 100%
dark-mode bool true 50%
pricing-v3 str "exp" 25%
legacy-api bool false
ai-copilot bool true 100%
$ brew install flagdash/tap/flagdash
j k nav c create t toggle / search q quit

Ready to ship faster?

Join teams who deploy with confidence using FlagDash. Set up in under 5 minutes.