Skip to content

anbeime/skill / skills/srt-whiteboard-animation

srt-whiteboard-animation

Verified

将 SRT 字幕做成暖米黄纸张底的白板手绘动画:读字幕→输出配图策略→确认后生成统一风格线稿→按叙事语义标注分区→预览台调整→渲染 MP4。编排沿用分区遮罩揭示(annotation.json / sequence / startMs / protectedRegions),但每个区域内的落墨换成 stream 的连续笔迹(骨架/网格 ink→color)。当用户提供 SRT 字幕并要求"字幕做成白板手绘/流式笔迹视频""SRT 生成白板动画""按字幕分镜画手绘"时触发。

scripts/prepare_env.py

View on GitHub
#!/usr/bin/env python3
"""
流式笔迹动画 - 环境引导脚本

职责:
  1. 在 skill 目录下建立隔离的 Python 虚拟环境(已存在则复用)
  2. 核对运行所需的第三方库是否可导入
  3. 自动补齐缺失的库
  4. 末行打印 ENV_PY=<解释器路径>,供上层调用方捕获

用法:
  python prepare_env.py          # 建环境 + 补依赖,输出 ENV_PY
  python prepare_env.py --check  # 仅探测,缺东西就以非零码退出
"""
from __future__ import annotations

import os
import subprocess
import sys
import venv
from pathlib import Path

# skill 根目录 = 本脚本向上两级
SKILL_ROOT = Path(__file__).resolve().parent.parent
VENV_ROOT = SKILL_ROOT / ".venv"

# 解释器导入名 -> pip 安装名
DEPS: dict[str, str] = {
    "cv2": "opencv-python",
    "numpy": "numpy",
    "av": "av",  # PyAV:纯 pip 安装的 H.264 编码,无需系统 ffmpeg
    "PIL": "Pillow",  # render_annotation_preview.py 画区域编号预览图(含中文标签)
}


def interpreter_path() -> Path:
    """虚拟环境里的 python 可执行文件位置(跨平台)。"""
    if sys.platform.startswith("win"):
        return VENV_ROOT / "Scripts" / "python.exe"
    return VENV_ROOT / "bin" / "python"


def ensure_venv(check_only: bool) -> Path:
    py = interpreter_path()
    if VENV_ROOT.exists() and py.exists():
        print(f"[ok] 复用现有虚拟环境: {VENV_ROOT}")
        return py

    if check_only:
        print(f"[err] 虚拟环境尚未建立: {VENV_ROOT}")
        sys.exit(1)

    print(f"[..] 建立虚拟环境: {VENV_ROOT}")
    venv.create(str(VENV_ROOT), with_pip=True)
    print("[ok] 虚拟环境就绪")
    return py


def can_import(py: Path, import_name: str) -> bool:
    probe = subprocess.run(
        [str(py), "-c", f"import {import_name}"],
        capture_output=True,
    )
    return probe.returncode == 0


def install(py: Path, packages: list[str]) -> bool:
    if not packages:
        return True
    print(f"[..] 安装依赖: {', '.join(packages)}")
    res = subprocess.run(
        [str(py), "-m", "pip", "install", "--quiet", *packages],
        capture_output=True,
        text=True,
    )
    if res.returncode != 0:
        print(f"[err] 安装失败:\n{res.stderr}")
        return False
    print("[ok] 依赖安装完成")
    return True


def main() -> None:
    check_only = "--check" in sys.argv

    py = ensure_venv(check_only)

    missing: list[str] = []
    for import_name, pip_name in DEPS.items():
        if can_import(py, import_name):
            print(f"[ok] {pip_name}")
        else:
            print(f"[miss] {pip_name}")
            missing.append(pip_name)

    if missing:
        if check_only:
            print(f"\n缺 {len(missing)} 个依赖: {', '.join(missing)}")
            sys.exit(1)
        if not install(py, missing):
            sys.exit(1)

    # 末行:供调用方捕获的约定输出
    print(f"\nENV_PY={py}")


if __name__ == "__main__":
    main()
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall