More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    VectorForge MCP Starter — Local SVG Asset & Export Workflow

    1

    Verifies the MCP server connection and bound project directory (`vectorforge_project_info`, or initializes via `vectorforge_init_project`).

    Free

    Security scanned
    VectorForge MCP Starter — Local SVG Asset & Export Workflow

    VectorForge MCP Starter — Local SVG Asset & Export Workflow

    Example session with this skill installed

    Create a new 'settings' icon variant for the 'dashboard-ui' family using the default project style, then give me a PNG snapshot.

    • Read your context and instructions
    • Compiled the vectorforge mcp starter
    • Generated the image

    I have connected to the VectorForge MCP server and identified the dashboard-ui family constraints. I authored a new settings icon using the approved #2D3436 and #0984E3 palette, ensuring the 24x24 viewbox matches existing sibling assets. The asset was created, validated against the project schema, and a deterministic PNG snapshot was rendered.

    Asset Generation Summary

    Asset IDStatusValidationSnapshot Path
    settings-icon-v2Created✅ Valid./snapshots/settings-icon-v2.png
    {
      "id": "settings-icon-v2",
      "name": "Settings Icon",
      "family": "dashboard-ui",
      "variant": "outline",
      "width": 24,
      "height": 24,
      "viewBox": [0, 0, 24, 24],
      "schemaVersion": 1,
      "svg": "<svg xmlns=\"http://www.w3.org/2000/svg\" width=\"24\" height=\"24\" viewBox=\"0 0 24 24\"><g id=\"main-path\" fill=\"none\" stroke=\"#2D3436\" stroke-width=\"2\"><path d=\"M12 15a3 3 0 100-6 3 3 0 000 6z\"/><path d=\"M19.4 15a1.65 1.65 0 00.33 1.82l.06.06a2 2 0 11-2.83 2.83l-.06-.06a1.65 1.65 0 00-1.82-.33 1.65 1.65 0 00-1 1.51V21a2 2 0 11-4 0v-.09a1.65 1.65 0 00-1-1.51 1.65 1.65 0 00-1.82.33l-.06.06a2 2 0 11-2.83-2.83l.06-.06a1.65 1.65 0 00.33-1.82 1.65 1.65 0 00-1.51-1H3a2 2 0 110-4h.09a1.65 1.65 0 001.51-1 1.65 1.65 0 00-.33-1.82l-.06-.06a2 2 0 112.83-2.83l.06.06a1.65 1.65 0 001.82.33 1.65 1.65 0 001-1.51V3a2 2 0 114 0v.09a1.65 1.65 0 001 1.51 1.65 1.65 0 001.82-.33l.06-.06a2 2 0 112.83 2.83l-.06.06a1.65 1.65 0 00-.33 1.82 1.65 1.65 0 001.51 1H21a2 2 0 110 4h-.09a1.65 1.65 0 00-1.51 1z\"/></g></svg>"
    }
    

    Next steps

    • Inspect the visual snapshot at ./snapshots/settings-icon-v2.png to confirm stroke alignment.
    • Run vectorforge_export with format: "web" to generate the ES module runtime.
    • Call vectorforge_create_variant if you need a 'hover' or 'active' state for this icon.

    settings-icon-v2.png

    PNG · 1536×1024

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    VectorForge MCP Starter turns your AI coding agent into a structured, style-aware 2D vector asset assistant powered by the open-source VectorForge pipeline and its official Model Context Protocol (MCP) stdio server.

    Instead of generating isolated, inconsistent raw SVGs in chat, an agent equipped with VectorForge MCP Starter follows a disciplined 11-step engineering loop directly against your local project files:

    1. Verifies the MCP server connection and bound project directory (vectorforge_project_info, or initializes via vectorforge_init_project).
    2. Reads project context, rules, naming conventions, and folder collections (vectorforge_project_context).
    3. Loads your target style.json palette, prohibited colors, and stroke rules (vectorforge_get_style).
    4. Inspects existing sibling SVGs (vectorforge_get_asset) to match proportions, silhouettes, and viewBox geometry.
    5. Authors self-contained, XML-safe SVG and schemaVersion: 1 metadata.
    6. Writes new assets (vectorforge_create_asset) or family-inheriting variants (vectorforge_create_variant) with atomic rollback protection.
    7. Validates schemas, cross-references, and objective style rules (vectorforge_validate).
    8. Renders deterministic 800×640 PNG snapshots (vectorforge_snapshot) for visual inspection.
    9. Evaluates the rendered PNG (when the agent supports vision) and repairs clipping or palette drift.
    10. Exports portable asset bundles (vectorforge_export) for Generic, Web, or Flutter targets using safe dry-run preflight checks.

    Key Features

    • Complete 11-Step Starter Workflow: Covers context discovery, style inspection, sibling analysis, asset/variant creation, validation, snapshot rendering, and dry-run exports.
    • Official MCP Tool Mapping: Aligned 1:1 with all 31 vectorforge_* MCP tools.
    • Multi-Client Setup Guides: Ready-to-use configurations for Claude Code, Cursor, VS Code / GitHub Copilot, and generic stdio MCP clients.
    • Diagnostic Helper Script: Includes scripts/verify-mcp.mjs to check Node.js compatibility, project initialization status, and MCP configuration.
    • 100% Local & MIT Open Source: Zero cloud accounts, zero telemetry, and no external API keys required inside the MCP server.

    Requirements

    • Operating Systems: Cross-platform — runs locally on Windows, macOS, and Linux.
    • Node.js 22+ (macOS, Linux, or Windows)
    • VectorForge CLI / MCP Server: Open-source (https://github.com/Kaandonmez/vector-forge or npm install -g vectorforge)
    • Optional: Playwright Chromium (npx playwright install chromium) for headless PNG snapshots

    Requires Node.js 22+, a running VectorForge MCP server, and Playwright Chromium for snapshots. Flutter exports lack a Dart runtime. Only supports SVG 2D assets.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 10 days ago

    • Free with an Agensi account

    Requires

    Windows
    macOS
    Linux
    Node.js 22+
    VectorForge MCP
    Vectorforge Starter MCP
    playwright
    Node.js

    Permissions

    Browser
    Read Files
    Write Files
    Network Access
    Listed10 days ago

    What's inside

    Frequently Asked Questions

    This skill enforces strict style governance (palettes, stroke widths, and geometry) and performs structural validation to prevent illegal XML tags and inconsistent viewboxes that standard LLMs often produce.

    The skill is designed for use with any AI agent that supports the Model Context Protocol (MCP), such as Claude Desktop or specialized IDE agents, provided they can connect to a local Node.js environment.

    You will need Node.js 22 or higher and a local VectorForge MCP server instance; if you plan to generate PNG snapshots for visual QA, a local Playwright Chromium installation is also required.

    Yes, the skill includes a multi-target export feature that bundles your validated assets specifically for Web (ES modules), Flutter projects, or generic SVG distributions.

    While the skill provides the asset bundle for Flutter, it does not include a native Dart animation runtime, and all PNG snapshot features require a local Playwright setup to function.

    This starter kit includes the core MCP skill definitions for style governance, asset generation, validation, and multi-target exports, allowing your agent to interface seamlessly with your local vector pipeline.

    New in Game Development