Integrations

Works Where You Work

Connect Pelaris to your favourite AI assistant. Your coach goes where you go.

Available Platforms

ChatGPT

Available on ChatGPT

Add Pelaris as a connector in ChatGPT. Ask your coach about your training, log workouts, and get personalised advice, all within your ChatGPT conversation.

Claude

Available on Claude

Connect Pelaris to Claude as an MCP integration. Get deep training analysis, thoughtful coaching conversations, and data-driven recommendations from your training history.

MCP Compatible

Works with any MCP client

Pelaris implements the open Model Context Protocol standard. Any MCP-compatible AI client can connect, including Cursor, Windsurf, Claude Code, and more.

Capabilities

What You Can Do

21 tools that give your AI assistant full access to your coaching engine.

Read Tools

  • • View training programs and session details
  • • Check benchmarks and personal records
  • • Get coaching insights and training context
  • • Search training resources and methodology guides
  • • Review body analysis and composition data

Write Tools

  • • Log workouts with sets, reps, and weights
  • • Create and modify planned training sessions
  • • Record benchmarks and track goals
  • • Submit daily check-ins and readiness scores
  • • Log injuries and swap exercises

AI Coaching

  • • Get personalised advice based on real training data
  • • Ask about recovery, fatigue, and readiness
  • • Generate weekly training plans
  • • Receive coaching feedback on completed sessions
  • • Get exercise alternatives and progressions

Get Started

How to Connect

Up and running in under a minute. No API keys, no configuration files.

1

Open your AI assistant

Head to ChatGPT or Claude in your browser.

2

Go to Settings

Navigate to Settings → Connectors (Claude) or Settings → Apps (ChatGPT).

3

Add Pelaris

Search for “Pelaris” or add the server URL directly:

https://api.pelaris.io/mcp
4

Sign in

Authorise with your Pelaris account. You control which scopes to grant.

5

Start training with AI

Ask your coach anything. “What’s my session today?” “Log my bench press.” “How should I adjust after missing last week?”

For Developers

Build on Pelaris

Standard MCP protocol. Standard OAuth. No proprietary SDK required.

Server Details

MCP Server URL

https://api.pelaris.io/mcp

Authentication

OAuth 2.0 with PKCE

Client Registration

Dynamic (RFC 7591)

Tools Available

21 tools (9 read, 12 write)

Protocol

Model Context Protocol

Privacy by Design

  • Pseudonymous user IDs — your real identity is never exposed to AI models
  • PII scrubbing on all data before it reaches any AI provider
  • Granular OAuth scopes — approve only what you need
  • Disconnect anytime — tokens revoked immediately

Your Data, Your Control

OAuth 2.0 authorisation with granular scope approval means your AI assistant only accesses what you explicitly allow. No background data sharing. No training on your data.

Disconnect at any time from your Pelaris settings or directly from your AI assistant. All tokens are revoked immediately.

Read our Privacy Policy →

OAuth 2.0 + PKCE authorisation flow

Granular scope control per connection

Pseudonymous IDs shield your identity

PII automatically scrubbed from AI context

No data shared without explicit consent

Instant token revocation on disconnect

Ready to train smarter?

Connect Pelaris to your AI assistant and start coaching conversations backed by real data.

Start Training →