No. Because MCP is an open standard, the same Harmny server works with any MCP-compatible client. Claude is the most common client today — see our Claude integration page for step-by-step setup.

⚡ Open standard · Model Context Protocol

Harmny speaks MCP — the protocol by Anthropic

The Model Context Protocol, introduced by Anthropic, is the open standard AI assistants use to connect to your tools and data. Harmny ships an MCP server — so your people and work context can flow into any MCP-compatible assistant, securely.

Set up with Claude Get started free

Read-only · Scoped API keys · Included on Growth and above

What the MCP server exposes

Six read-only capability areas, each gated by your API key’s scopes.

Career frameworks & competencies

Read every framework and its competencies.

Career paths

Levels and skill descriptions for any framework.

Employee summaries

Ratings, goals, and current level for a person.

Org overview

High-level people metrics for the organization.

Tasks

Filtered by project, assignee, status, or priority.

Projects & spaces

Spaces and their active projects.

For developers

Transport: Streamable HTTP (POST + SSE) at /api/mcp
Discovery: /.well-known/mcp/server-card.json
Auth: Scoped API key, per-key rate limiting, audit logging
Access: Read-only, isolated per organization

Ready to connect? The Claude integration guide walks through key generation and client setup step by step.

MCP & Anthropic FAQ

What is the Model Context Protocol (MCP)?

How does Harmny implement MCP?

Is it Claude-only?

Does Anthropic train on our data?

Is the connection read or write?

Clarity. Motivation. Growth. For your whole team.

Free forever for up to 10 users. No credit card required. Full access from day one.

Get started free Book a 20-minute demo

By signing up you agree to our Terms of Service and Privacy Policy.