What is Toqua MCP?

Toqua MCP lets AI agents securely access Toqua on your behalf.

This means you can ask questions about vessel performance in natural language instead of manually finding and combining information across dashboards, models, and APIs.

For example, you can ask:

“How has Vessel A's hull performance evolved over the past year?”

“What would Vessel A consume at 12.5 knots in these conditions?”

The agent reasons and retrieves information from Toqua to answer the question.

What can you use it for?

Monitoring hull performance

Understand how a vessel's hull performance evolves over time, identify degradation, and evaluate the effect of cleaning or drydocking.

Try asking:

“How has Vessel A's hull performance evolved since its last drydock?”

“What did the last cleaning save us, and does it look like we should clean again?”

This can help support decisions around hull and propeller cleaning, quantify the impact of previous interventions, and identify vessels where degradation deserves further investigation.

Benchmark vessels across your fleet

A single vessel's performance is often more useful when put into context.

You can use Toqua MCP to compare performance across your fleet or between similar vessels and identify which vessels are deteriorating faster or performing differently from their peers.

Try asking:

“How does Vessel A's performance compare with its sister vessels?”

“Which vessels in my fleet show the most hull degradation?”

You can then drill into an individual vessel to understand why it stands out.

Explore speed–fuel performance

Use Toqua's vessel-specific performance models to evaluate the relationship between speed, fuel consumption, power, RPM, draft, and environmental conditions.

Rather than relying on a single static speed-consumption table, you can ask questions for the conditions that matter to you.

Try asking:

“What would Vessel A consume at 12.5 knots, laden, in Beaufort 4?”

“Show me the fuel impact of sailing between 11 and 14 knots.”

Set and monitor performance warranties

Use current vessel performance when setting charter-party speed and consumption warranties, and monitor commercial exposure once a vessel is on charter.

Before fixing, you can test whether a proposed warranty leaves sufficient margin against the vessel's expected performance.

Once the vessel is on charter, you can compare its modelled performance against the warranted conditions and investigate potential claim exposure.

Try asking:

“I want to TC out Vessel A. What speed-consumption warranty could I realistically offer?”

“Are we currently at risk of a performance claim on Vessel A?”

You can also ask follow-up questions about individual speeds, drafts, weather limits, contractual abouts, fuel margins, or estimated financial exposure.

Ask follow-up questions

You do not need to formulate the entire analysis in one prompt.

For example, you might start with:

“Which vessels show signs of fouling?”

Then continue with:

“Tell me more about Vessel A.”

“What did its last cleaning save?”

“How does it compare with its sister vessels?”

“We're considering chartering it out. What should we warrant?”

The assistant keeps the context of the conversation and can retrieve additional information from Toqua as the question develops.

---

Connecting Toqua

Connecting with Microsoft Copilot (Copilot Studio)

Setting up the MCP server is done by creating an agent in Copilot Studio. Afterwards the agents can be added to Copilot Chat and Microsoft Teams, as well as shared with colleagues.

Copilot Studio has recently received a visual update. If your interface looks the same as the video below, you're using the new experience. If not, please follow the text instructions further below.

Old Copilot Studio experience

Below instructions are only applicable if you're on the old Copilot Studio experience.

1. Add the MCP server

1. In Copilot Studio, select the environment you want to work in.
2. Open your agent (or create a new one) and go to the Tools page.
3. Select Add a tool, then New tool, then Model Context Protocol.
4. In the onboarding wizard, fill in:
   • Server name — e.g. Toqua
   • Server description — e.g. Monitor your fleet's performance
   • Server URL — https://api.toqua.ai/mcp
   • Authentication — OAuth 2.0
     • Type — Dynamic Discovery
5. Complete the wizard. Copilot Studio creates a custom connector behind the scenes that routes requests to the server over Streamable HTTP.

Note: The MCP onboarding wizard, which handles OAuth discovery automatically, is still rolling out and may not be enabled for every tenant or region. If selecting Model Context Protocol only opens a documentation page instead of the wizard, your tenant likely hasn't received the feature yet. In that case, open a Microsoft support ticket to check your rollout status — the manual "custom connector" path does not support automatic OAuth discovery.

2. Authorize and connect

1. The first time the connection is used, you'll be prompted to authenticate. Select Create connection and sign in.
2. You'll be redirected to authorize Toqua access via OAuth. Approve the request to complete the connection.

3. Add the server to your agent

1. On the Tools page, select Add a tool, then Model Context Protocol.
2. Select the Toqua connector from the list.
3. Select Add and configure. The Toqua tools become available to your agent.

You can now ask the agent questions about your fleet, and it will use the Toqua tools to answer.

Connecting with ChatGPT

In ChatGPT, the Toqua MCP server is added as a custom app. A ChatGPT Plus account is required. Free accounts do not support adding custom MCP servers.

1. First, make sure Developer Mode is enabled. Go to Settings → Security and login and enable Developer Mode.
2. Open Plugins, click the + button, and enter as connection name "Toqua MCP".
3. Enter the server URL: https://api.toqua.ai/mcp . Leave the other settings to their defaults.
4. Click Create and then Sign in with Toqua MCP. You'll be redirected to sign in with your Toqua account.
5. Create a new chat and start using it - e.g. ask "What is the current hull performance of my fleet?".

Connecting with Claude

In Claude, the Toqua MCP server is added as a custom connector. It works across Claude, Claude Desktop, and the mobile apps, on Pro, Max, Team, and Enterprise plans.

On Team and Enterprise plans, only an Owner can add the connector via Organization Settings → Connectors. Once added, each member connects and authorizes individually. On Pro and Max, you add it directly in your own settings.

1. Go to Settings → Connectors (or Customize → Connectors).
2. Click the + next to Connectors, then Add custom connector.
3. Enter a name (e.g. Toqua) and the server URL: https://api.toqua.ai/mcp. You can leave the advanced OAuth settings blank — Toqua handles registration automatically.
4. Click Add, then Connect, and sign in to authorize access via OAuth.

Connecting with other agents

If you wish to connect Toqua MCP to any other services, below settings should be used.

Authentication is handled through OAuth. Upon first connection, you'll be redirected to sign in and authorize access.