Hyphen MCP Server
Connect Claude, Cursor, VS Code, or any MCP client to your Hyphen organization — deploy apps, manage feature flags, investigate incidents, and operate automations from your AI tools, with the same permissions you have in the app.
Connect your client in minutes
Hyphen hosts the server — there is nothing to install or run yourself. Point any MCP client at the endpoint and sign in with OAuth, or use an API key from the Hyphen app.
One command — Claude Code signs you in with OAuth the first time it connects:
claude mcp add --transport http hyphen https://mcp.hyphen.ai/
Open source at github.com/Hyphen/mcp and published to the official MCP registry as ai.hyphen/mcp.
60+ tools across the platform, one connection
Ship From a Conversation
The packaged deploy_app workflow analyzes the repo your coding agent has open, sets up the project, app, and Docker packaging, ships preview-first, watches the run and logs, and promotes to production only after you confirm.
Manage Feature Flags End to End
Create and update toggles and segments, adjust targeting, and verify with a live edge evaluation that returns the same answer a real user would receive.
Clean Up Stale Flags
Telemetry-backed tools find flags that no longer vary, and a guided workflow removes them from your codebase before retiring them in Hyphen.
Investigate Incidents
Search the organization audit log and correlate it with deployment runs, logs, and metrics to answer “what changed right before things broke?”
Operate Agent Tasks
Schedule and inspect Hyphen Agent automations, review their runs, and answer their questions when a run is waiting on input.
Links, Geolocation & Live Docs
Create short links and QR codes, analyze click performance, look up IP geolocation, and answer platform questions from live documentation instead of stale training data.
Built for safe autonomy
Agents get real capability, and you keep control — enforced by the Hyphen API on every request, not by trust in the model.
Your Credential, Your Permissions
Every tool call runs as you. What a connected agent may do is decided per request from your organization membership and roles — the same rules as the Hyphen App.
Permission Prompts on Writes
Tools that modify data are annotated as such, and deletes, deployment triggers, and agent chat carry a destructive hint, so MCP clients ask for your confirmation before running them.
A Production Gate on Deployments
A deployment run that targets a production-type environment is refused unless the call explicitly acknowledges production — and agents are instructed to do that only after you confirm.
Secrets Stay Encrypted
ENV values are end-to-end encrypted with a key only you hold, so the server can only ever read secrets metadata — names, counts, versions — never values.