v0.2.0

v0.2.0 (2026-05-26)

Minor release: Caddy-based Docker deployment with standard HTTP/HTTPS ports, optional Let's Encrypt for public domains, and several frontend improvements.

Highlights

• Caddy reverse proxy — Replaces nginx in the frontend container; serves static assets and proxies /api and /ws to the backend.
• Standard ports — Docker UI at http://localhost (port 80). HTTPS on port 443 when using a public domain.
• Public HTTPS — Set VDS_DOMAIN and optional ACME_EMAIL in .env for automatic Let's Encrypt certificates.
• Frontend — Vision provider presets in Playground, unified AppSelect dropdowns, IBM Plex typography, home page polish, pointer cursor fixes.

Changed

• Removed FRONTEND_PORT; host mapping is fixed to 80:80 and 443:443.
• Install / run scripts resolve the UI URL from .env (http://localhost or https://<VDS_DOMAIN>).
• run-in-docker.cmd uses curl.exe on Windows for readiness checks.

Install (this release)

macOS / Linux

curl -fsSL https://raw.githubusercontent.com/thought2code/video-driven-skill/v0.2.0/scripts/install.sh | bash -s -- --ref v0.2.0 --tag v0.2.0

Windows (PowerShell)

$env:VD_SKILL_REF='v0.2.0'; $env:VD_SKILL_IMAGE_TAG='v0.2.0'; irm https://raw.githubusercontent.com/thought2code/video-driven-skill/v0.2.0/scripts/install.ps1 | iex

Set AI_API_KEY in .env, then open http://localhost.

For a public domain with HTTPS, add to .env:

VDS_DOMAIN=vds.example.com
ACME_EMAIL=you@example.com

Ensure DNS points to your server and ports 80 / 443 are reachable from the internet.

Container images

Image	Tags
ghcr.io/thought2code/video-driven-skill-backend	v0.2.0, v0.2, latest
ghcr.io/thought2code/video-driven-skill-frontend	v0.2.0, v0.2, latest

Images are built automatically by CI when this tag is pushed.

Requirements

• Docker
• OpenAI-compatible API key (AI_API_KEY in .env) with a vision-capable model for skill generation

v0.1.1

v0.1.1 (2026-05-20)

Patch release: align GHCR image registry and install script URLs with the repository owner thought2code.

Fixed

• docker-compose.release.yml — image paths now use ghcr.io/thought2code/...
• scripts/install.sh / scripts/install.ps1 — download compose and .env from thought2code/video-driven-skill
• README.md / README.zh-CN.md — install docs and GHCR links updated

Install (this release)

macOS / Linux

curl -fsSL https://raw.githubusercontent.com/thought2code/video-driven-skill/v0.1.1/scripts/install.sh | bash -s -- --ref v0.1.1 --tag v0.1.1

Windows (PowerShell)

$env:VD_SKILL_REF='v0.1.1'; $env:VD_SKILL_IMAGE_TAG='v0.1.1'; irm https://raw.githubusercontent.com/thought2code/video-driven-skill/v0.1.1/scripts/install.ps1 | iex

Set AI_API_KEY in .env, then open http://localhost:3000.

Container images

Image	Tags
ghcr.io/thought2code/video-driven-skill-backend	v0.1.1, latest
ghcr.io/thought2code/video-driven-skill-frontend	v0.1.1, latest

v0.1.0

Video Driven Skill v0.1.0 (2026-05-20)

First versioned release with pre-built Docker images on GitHub Container Registry (GHCR) and a no-clone install path for end users.

Note: Compose files and install scripts in this tag still reference ingorewho for GHCR. Use v0.1.1 for a working out-of-the-box install, or apply the registry fix manually when using v0.1.0 images.

Highlights

• GHCR images — Backend and frontend images are built and published automatically when a v* Git tag is pushed.
• Release install — docker-compose.release.yml plus scripts/install.sh / scripts/install.ps1 for running without cloning the repo.
• Documentation — README (EN / 简体中文) reorganized with separate paths for end users vs developers.

Container images

Image	Tag examples
ghcr.io/thought2code/video-driven-skill-backend	v0.1.0, latest
ghcr.io/thought2code/video-driven-skill-frontend	v0.1.0, latest

Quick start (Docker)

Recommended: install v0.1.1 instead.

For this tag’s images only:

export VD_SKILL_IMAGE_TAG=v0.1.0
docker pull ghcr.io/thought2code/video-driven-skill-backend:v0.1.0
docker pull ghcr.io/thought2code/video-driven-skill-frontend:v0.1.0

Configure via build from source or upgrade to v0.1.1 install scripts.

Requirements

• Docker
• An OpenAI-compatible API key (AI_API_KEY in .env) for skill generation