
Build log
Explore model projects, build stages, materials, decisions, and technical notes.
Search Glushkov Modelling’s published wooden ship-model archive with Perplexity, Claude, ChatGPT, Cursor, or another compatible AI assistant.
Ask a connected assistant to find published material across build logs, model research, gallery photographs, and practical tutorials.

Explore model projects, build stages, materials, decisions, and technical notes.

Find model scale, manufacturer, build status, and historical context.

Browse photographs of hulls, decks, rigging, fittings, and finished models

Search educational articles about techniques, materials, tools, and modelling practice.
Choose your assistant, authorize the connection, then ask questions in the same way you would ask a knowledgeable workshop companion.
Select one of the available AI platforms and open its connection guide.
Add the public MCP endpoint and complete the OAuth connection in the selected assistant.
Search build logs, model information, gallery photographs, and tutorials through the connected assistant.
Select a compatible client, then follow its connection guide to add Glushkov Modelling through the public OAuth MCP endpoint.

Perplexity is an AI-powered answer engine designed for research and fact-finding. It searches the web in real time, synthesizes information into concise responses, and typically provides citations so users can trace claims back to original sources.

Claude is an AI assistant developed by Anthropic, known for thoughtful long-form reasoning, clear writing, document analysis, and coding support. It is often used for drafting, summarizing large amounts of text, exploring ideas, and handling complex conversational tasks.

ChatGPT is an AI assistant developed by OpenAI that helps users write, brainstorm, learn, analyze information, generate code, and work with files and images. It is designed as a flexible general-purpose conversational tool for both everyday and professional tasks.

Cursor is an AI-first code editor built around software development workflows. It integrates AI assistance directly into the programming environment, helping developers understand codebases, generate or modify code across files, debug issues, and work more efficiently with natural-language instructions.

Visual Studio Code with GitHub Copilot can connect to remote MCP servers and make their tools available in Copilot Chat.

Google Antigravity is an agentic development environment that can connect to MCP servers and use their tools during research and coding workflows.
The official server is published in selected public MCP directories. Choose a directory to inspect the listing or install the connection.
Use these examples after connecting Glushkov Modelling to a compatible AI assistant. You can copy a prompt as written or adapt it to your own project.
Find the build-log stages for Le Requin 1750 and show links to the original source pages.
Le Requin 1750: Build-Log Stages
A completed 1:48 ZHL Model build of the French Mediterranean xebec.
The Le Requin project is documented in six chronological build-log stages. The sequence moves from constructing the building slip and hull framework to deck work, fittings, spars and sails, and the final finishing details.
Recorded 1 February 2025
The project begins with a custom plywood building slip. Its flat, stable structure provides the reference foundation on which the model is assembled.
Recorded 19 October 2025
The hull’s structural framework is assembled: keel, frame set, stem, and sternpost form the ship’s basic skeleton.
Recorded 22 November 2025
Work moves to deck planking and the armament. The main deck is laid first, including reinforcement in the areas of the future mainmast and cargo-hatch coaming.
Recorded 6 February 2026
This stage documents construction of the ship’s boat, using a more detailed method than a simple painted casting or solid carved substitute.
Recorded 6 March 2026
Masts and yards are made from laminated square blanks and gradually shaped with a plane and lathe. The work then proceeds to the sail plan.
Recorded 26 March 2026
The final stage covers delicate completion work, including correctly laid anchor cables, partially secured xebec oars, and final finishing details.
All six stages are collected on the Le Requin project page, which also includes the project’s historical note and model specifications.
What are the scale, manufacturer, and current build status of HMS Bounty?
According to the HMAV Bounty project record, the model is built to 1:60 scale, the kit manufacturer is Amati, and its current build status is completed
Summarise the historical context of the prototype behind Le Requin 1750. Include links to the relevant source materials.
Le Requin (c. 1750): Historical Context
The prototype behind the 1:48 ZHL Model interpretation
Le Requin—“The Shark”—was a French Royal Navy Mediterranean xebec built at Toulon around 1750–1751. It belonged to France’s light naval forces under Louis XV: a type valued for speed, shallow-water utility, manoeuvrability, and the ability to combine sailing with oar propulsion.
The xebec was particularly suited to the Mediterranean theatre. Its narrow, elongated hull, strongly projecting bow and stern, and lateen-rigged masts expressed a design priority very different from that of a conventional ship of the line: rapid movement, agile handling, and effectiveness close to coasts. For a navy operating from Toulon, these were useful qualities in waters defined by busy sea lanes, island passages, coastal approaches, and the persistent threat of corsair activity.
Le Requin could both sail and row. This hybrid capability gave her practical flexibility when wind conditions were poor and made the vessel well suited to short-range patrol, chase, and reconnaissance duties.
Built for the French Royal Navy, Le Requin served in the Mediterranean for more than a decade before withdrawal from service in the 1760s. Her reported duties included:
These tasks explain why a fast, lightly armed vessel was strategically valuable. Instead of relying on heavy broadside power, a xebec could identify threats, carry intelligence, protect vulnerable traffic, and pursue or evade smaller opponents in the confined and changeable conditions of the Mediterranean.
The model’s distinctive visual features—the slender hull, pronounced bow and stern overhangs, and lateen rig—are not merely decorative. They reflect the prototype’s function as a light, fast, manoeuvrable warship. The build log documents how those characteristics are interpreted at 1:48 scale, from the hull framework through the deck and armament to spars, sails, rigging, oars, and final details.
Find gallery photographs that show deck fittings on HMAV Bounty.
HMAV Bounty: Deck Fittings Gallery
Two gallery photographs are tagged deck for the HMAV Bounty project. Together, they provide useful broad views of the boat deck and main deck, where the arrangement and relationship of deck fittings can be studied in context.
Open the HMAV Bounty build page · Open the complete Bounty gallery
Find tutorial articles about hull planking suitable for a beginner, and briefly explain which one to start with.
Beginner Hull Planking Tutorials
Start with “Hull Planking Basics”: it is the core introductory article on double hull planking—first creating a rough, fair base layer, then applying the finish planking over it. Follow it with “Wood Bending” so you can shape planks confidently around the curved bow and stern sections.
| Article | Level | What It Gives a Beginner | When to Read It |
|---|---|---|---|
| Hull Planking Basics | Beginner | Introduces the double-planking method: the purpose of the initial rough layer and the subsequent finish planking. | First |
| Wood Bending | Beginner | Explains how to give wooden parts—including hull planks—the required curve when simply fastening them to the framework is not enough. | Second, before working on the bow and stern. |
The site currently has 17 tutorial articles. The Planking category contains four articles, two of which are marked Beginner; these are the most appropriate starting point for learning hull planking.
This article establishes the correct working sequence:
For an approach that uses soft limewood strips for the first layer, this sequence is especially logical: limewood is easier to fit along complex hull curves, while the first layer provides a firm, sandable substrate for the finish planking. The wood-bending article then helps you pre-form a plank at the stem rather than forcing it into position and risking splits or unfair curves.
Add the public endpoint in the MCP or integrations section of your selected AI assistant, then complete OAuth authorization when prompted.
https://mcp.glushkov-modelling.com/mcp-oauthUse an assistant that supports MCP connections and OAuth authorization.
Quick guidance for the most common OAuth, endpoint, and tool discovery issues.
If your MCP client cannot complete registration, its OAuth redirect URI may not be approved by security policy. To connect successfully, your client should use a standard loopback callback, such as [http://127.0.0.1:<port>/callback] or [http://localhost:<port>/callback] For clients that use a public HTTPS callback URL, please contact me ([email protected]) with [ 1. The name and version of the MCP client 2. The exact redirect_uri requested during registration 3. A link to the client developer’s official documentation] After verification the callback domain, I can safely allow it for all users of that MCP client. I do not automatically approve unknown callback domains.
Applies to: Perplexity, Claude, ChatGPT, Cursor, Visual Studio Code, Google Antigravity
Complete the OAuth flow in the same browser window opened by the client. Allow cookies and popup windows, then return to the client only after sign-in has completed.
Applies to: Perplexity, Claude, ChatGPT, Cursor, Visual Studio Code, Google Antigravity
Close and reopen the chat or application, then check the connector status in the client settings. Confirm that the connector is connected and enabled for the current chat or Agent session.
Applies to: Perplexity, Claude, ChatGPT, Cursor, Visual Studio Code, Google Antigravity
Short answers about the public MCP server, access, privacy, and supported archive material.
No. The public Glushkov Modelling MCP connection uses OAuth and automatic client registration. Add the public OAuth endpoint in a compatible client, then complete sign-in or registration in the browser window opened by that client.
AccessThis is expected. The MCP endpoint accepts programmatic JSON-RPC requests from an MCP client and does not serve a normal browser page through a GET request. Add the endpoint in a compatible AI client instead of opening it directly in a browser.
ConnectionNo. Glushkov Modelling MCP provides read-only access to published Glushkov Modelling archive content. It does not access your local files, private chats, or other data in your AI client.
PrivacyIf you need help connecting Glushkov Modelling MCP, encounter an OAuth or endpoint issue, or notice that a platform guide needs updating, please get in touch. Your feedback helps keep the connection guides accurate and useful.