samchon/blendia

Última actividad de origen registrada: 3 de noviembre de 2026 a las 07:21

Catálogo de SkillsMP actualizado: 12 de noviembre de 2026 a las 12:52

skills recopiladas: 7

Estrellas en GitHub: 0
Forks en GitHub: 0

Instalar con un asistente de IA

Copia este prompt en el asistente de IA que estés usando.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/samchon/blendia.

README del repositorio

Archivo de origen del README: README.md

Leer el README original

Mostrar README completo

# Blendia Blendia brings TypeScript scripting to an existing Blender installation. It is designed for people and coding agents working in TypeScript projects, with Blender handling modeling, scene evaluation, and rendering. Run a project script to create and inspect Blender scenes through a typed SDK, including meshes, materials, cameras, lights, native PNG renders, and `.blend` copies. The Blendia extension connects the script to the live scene. ## Status Blendia is in early development and is not yet published to npm. The current repository version supports saved project files; setup and usage are documented below. The Blender extension includes preview actions for running project files and active Text snapshots. Their setup and limitations are described in the [extension guide](blender/extension/README.md#run-typescript-from-blender). ## Run a project script Enable the [Blender extension](blender/extension/README.md) and configure its connection as described in the [CLI guide](packages/cli/README.md). Follow the [external-project setup](docs/design/script-jobs.md#set-up-an-external-source-checkout-project) to connect this checkout's unpublished SDK and prepare your project's compiler and dependencies. Blendia supplies the public `ttsx` launcher. From the repository with its dependencies installed, run the source CLI directly: ```sh pnpm blendia run --cwd /path/to/project --project tsconfig.json src/model.ts ``` See the [execution and diagnostics guide](docs/design/script-jobs.md#first-run-preparation) for first-run native compiler/plugin preparation, structured output, session discovery, and metadata-only setup checks. ## Run from Blender The extension's TypeScript Macros panel provides Run TypeScript File for a saved project entry and Run Active Text Snapshot for the selected Text Editor buffer. Both use your project's compiler and dependencies. Follow the [extension setup](blender/extension/README.md#run-typescript-from-blender) to select the project and execution tools. An active Text snapshot runs as a sibling module in an explicit project directory. The original disk file and Text are preserved, but module identity changes, self-imports still use the original file, and old on-disk errors can still fail project checking. There is no macro Stop or automatic rollback: stopping the bridge does not cancel the script, and changes already made to Blender can remain after failure. ## Usage documentation The current interfaces and their limitations are documented here: - [CLI](packages/cli/README.md): running saved project TypeScript files. - [SDK](packages/sdk/README.md): connecting to Blender and working with scene objects. - [MCP agent tools](docs/design/mcp-agent-tools.md): connecting a coding agent to scene tools, captured scripts, and bounded native PNG images. - [Scene authoring and outputs](docs/design/scene-render-save.md): creating a scene, rendering a PNG, and saving a `.blend` copy with explicit output ownership. - [Owned headless projects](docs/design/headless-projects.md): running one project script in a new background Blender with explicit executable and extension paths. - [Blender extension](blender/extension/README.md): establishing the live Blender connection. - [Observed environments and limits](docs/support.md): recorded Blender and agent-host workflows, plus first-use preparation limits.

Cargando el README del repositorio…

Skills en este repositorio

clasificación pendiente

skill

ocupación

descripción

actualizado

contracts

sin clasificar

Owns Blendia declaration and test Evidence checklists and their package selectors. Use when writing or reviewing selected declarations, test scenarios, or Evidence configuration.