Overview

The Go1 MCP (Model Context Protocol) server lets you connect AI agents and MCP-compatible tools directly to the Go1 learning catalogue. Once connected, your AI assistant can search for courses, explore available skills, surface learning playlists, and deliver personalised learning recommendations — all using natural language.

This article explains how to connect to the Go1 MCP server, what tools are available, and how to get the most out of the integration.

Before you begin

To connect to the Go1 MCP server, you will need:

An active Go1 account
A valid Go1 user JWT (JSON Web Token), obtained via your OAuth flow
An MCP-compatible client — such as Claude, Cursor, or an AI agent built with an MCP SDK

Note: Your portal's content entitlements apply to all MCP requests. You will only see content that your Go1 portal has access to.

Connecting your MCP client

MCP endpoint

Point your MCP client to the following URL:

https://mcp.go1.com/mcp

Authentication

All requests require a Go1 user JWT passed as a Bearer token in the Authorization header:

Authorization: Bearer <your_go1_user_jwt>

The server validates your token and enforces portal-level permissions automatically. We recommend storing your JWT in an environment variable rather than hardcoding it:

GO1_MCP_SERVER_URL=https://mcp.go1.com/mcp REQUEST_TIMEOUT=300

Tip: Validate that your JWT has more than 60 seconds of remaining validity before connecting to avoid failures mid-call.

Steps to connect

Obtain a Go1 user JWT via your OAuth flow.
Configure your MCP client with the endpoint URL and your JWT.
Complete the MCP handshake (tools/list) — your client will discover all available tools and their schemas automatically.
Start using natural language to search and explore Go1 learning content.

Available tools

Once connected, the following tools are available for content discovery.

search_content

Search the Go1 catalogue by keyword, with optional filters for duration, level, language, content provider, and skill. Returns course titles, descriptions, and direct links.

Example prompts:

"Find me leadership courses in English under two hours"
"What cybersecurity training is available for beginners?"
"Show me project management courses from LinkedIn Learning"
"Find compliance training available in Spanish"

get_content_facets

Returns a breakdown of available filters for a given search — including languages, content providers, and skill tags. Useful for understanding the search space before running a full search.

Example prompts:

"What languages is leadership training available in?"
"Which providers offer data analysis courses on Go1?"
"Show me the breakdown of available providers for Excel training"

suggest_skills

Returns skill recommendations based on a role title or topic. Use these as filter inputs when searching for learning content, to surface targeted results clustered by skill.

Example prompts:

"What skills should a data analyst develop?"
"What learning areas are relevant for a new manager?"
"Suggest skills for someone moving into a product role"

get_playlist_details

Retrieves the full ordered contents of a curated Go1 learning playlist, including titles, durations, providers, and links for each item in the sequence.

Example prompts:

"Show me what's included in the cybersecurity essentials playlist"
"What courses make up the new manager onboarding learning path?"
"List all items in the Go1 compliance fundamentals playlist"

get_content_alternatives

Suggests alternative learning content for a specific item that has been decommissioned. Pass the ID of the original item to find equivalent replacements.

Example prompts:

"Find alternatives to course ID abc123 in case it's been removed"
"Are there equivalent options to replace this deprecated training?"

Typical workflows

Finding courses for a topic

Call get_content_facets with your topic (e.g. "leadership") to understand the available languages and providers.
Call search_content with your chosen filters and a result limit of 10–20.
Optionally refine further using additional filter combinations.

Recommending skills for a role

Call suggest_skills with a role or topic (e.g. "data analyst").
Use the returned skills as filter inputs to search_content to find targeted learning content.
Present results grouped by skill to create a structured learning path.

Surfacing a curated playlist

Use search_content or a known playlist ID to locate the playlist.
Call get_playlist_details with the playlist ID to retrieve the full ordered list of items.
Return the sequence with metadata for each item.

Troubleshooting

Issue

Likely cause

What to try

Issue	Likely cause	What to try
`401 Unauthorized`	Missing or expired JWT	Refresh your token and reconnect. Check that the `Authorization` header is present and the token has not expired.
`403 Forbidden`	Portal access restriction	Confirm your portal's content entitlements. Contact your Go1 account team if you believe access should be granted.
Empty results	Filters too narrow, or portal entitlements	Try a broader keyword first, then use `get_content_facets` to understand what's available before narrowing down.
Response cut off	Response size limit exceeded	Reduce your `limit` parameter (10–20 is recommended) and request detailed fields selectively.
Timeout	Overly broad query or high load	Narrow your filters, reduce `limit`, or increase your client's `REQUEST_TIMEOUT` setting.

Security

Store your JWT in an environment variable or secrets manager — never in a plain config file or committed to source control.
Validate JWT expiry before connecting to avoid failures during a call.
If you are using multiple MCP servers alongside Go1, ensure your client namespaces tools correctly to avoid naming conflicts.

Getting the most from your Go1 Learn homepage

Curate Product Updates

Go1 Learn: Your Complete Admin Guide

Go1 Learn: Learner Guide