hashgraph-online/awesome-codex-plugins / plugins/andreamzhang/jimeng-skills/.agents/skills/jimeng-storyboard
jimeng-storyboard
VerifiedUse when 编写即梦分镜。按讲戏本剧情点映射母镜头、填7字段。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.1 KB |
| config/duration-rules.json | config | 2.1 KB |
| examples/jimeng-prompt-examples.md | markdown | 12.2 KB |
| scripts/apply_corrections.py | script | 13.6 KB |
| scripts/auto_correct.py | script | 4.7 KB |
| scripts/auto_header.py | script | 4.8 KB |
| scripts/calc_duration.py | script | 15.0 KB |
| scripts/check_assets_change.py | script | 9.4 KB |
| scripts/check_budget.py | script | 6.4 KB |
| scripts/check_dialogue.py | script | 11.6 KB |
| scripts/check_mother_id.py | script | 3.5 KB |
| scripts/check_picture_field.py | script | 8.8 KB |
| scripts/gen_clean.py | script | 2.3 KB |
| scripts/gen_dialogue_list.py | script | 10.3 KB |
| scripts/gen_episode_brief.py | script | 5.5 KB |
| scripts/run_all_checks.py | script | 7.6 KB |
| scripts/stage5_pipeline.py | script | 3.6 KB |
| templates/dialogue-shots.md | markdown | 6.3 KB |
| templates/jimeng-prompts-template.md | markdown | 7.8 KB |
scripts/check_budget.py
View on GitHub#!/usr/bin/env python3
"""
check_budget.py —— 讲戏本预算自检脚本
在导演阶段(阶段二产出后)校验导演讲戏本的时间预算是否自洽,从源头拦截
"预算只算台词、没算视觉动作预算"的系统性问题(如 ep53 讲戏本累加 53s vs 分镜实际 68s 差 15s)。
校验项:
1. 硬错误:讲戏本各 P 预算之和 < 台词本全片台词合计 → 台词塞不下(余量为负,必返工)。
2. 硬错误:某个 P 预算 > 15s → 违反母镜头 4-15s 硬约束。
3. 软提示:余量(预算之和 - 台词合计)异常偏高(> 高余量阈值)→ 可能为凑时长加戏,需人工确认。
4. 信息:输出 预算之和 / 台词合计 / 余量 / 头声明的总预算(可选)对照。
用法:
python check_budget.py --director <01-director-analysis.md> --dialogue <epXX-dialogue-list.md>
退出码: 0 = 通过, 1 = 存在硬错误, 2 = 用法错误/文件缺失
"""
import re
import sys
from pathlib import Path
# Windows GBK 控制台兼容:强制 UTF-8 输出,避免 ✓/✗/⚠ 触发 UnicodeEncodeError 崩溃
try:
sys.stdout.reconfigure(encoding='utf-8')
except Exception:
pass
# 母镜头总时长硬上限(秒)
MOTHER_MAX = 15.0
# 余量偏高阈值(秒):预算之和 - 台词合计 超过此值视为"可能为凑时长加戏"
HIGH_MARGIN = 20.0
def main():
args = sys.argv[1:]
director_path = None
dialogue_path = None
i = 0
while i < len(args):
if args[i] == '--director' and i + 1 < len(args):
director_path = args[i + 1]
i += 2
elif args[i] == '--dialogue' and i + 1 < len(args):
dialogue_path = args[i + 1]
i += 2
else:
i += 1
if not director_path or not dialogue_path:
print("用法: python check_budget.py --director <01-director-analysis.md> --dialogue <epXX-dialogue-list.md>")
sys.exit(2)
if not Path(director_path).exists():
print(f"错误: 讲戏本不存在: {director_path}")
sys.exit(2)
if not Path(dialogue_path).exists():
print(f"错误: 台词本不存在: {dialogue_path}")
sys.exit(2)
# ── 1. 讲戏本:提取各 P 的时长 ──
with open(director_path, encoding='utf-8') as f:
director = f.read()
p_budgets = []
# 兼容两种讲戏本格式:
# 旧格式:块内有 "- 时长: X秒"(deepseek 版 ep54)
# v2 精简格式:时长写进标题尾部 "## Pxx ...(Xs秒)",body 无"时长:"键(Codex 版 ep55)
for pm in re.finditer(r'^##\s*(P\d+)([^\n]*)\n(.*?)(?=^##\s*P\d+\b|^##\s+人物清单|^##\s+场景清单|^##\s+关键问题点汇总|\Z)', director, re.M | re.S):
pid, title_rest, body = pm.group(1), pm.group(2), pm.group(3)
tm = (re.search(r'时长\*\*[::]?\s*([\d.]+)\s*秒', body)
or re.search(r'时长[::]\s*([\d.]+)\s*秒', body)
or re.search(r'(\s*([\d.]+)\s*秒\s*)', title_rest))
if tm:
p_budgets.append((pid, float(tm.group(1))))
# 头部声明的总预算(可选信息)
header_total = None
htm = re.search(r'总时长\s*[|~::]\s*([\d.]+)\s*秒', director)
if htm:
header_total = float(htm.group(1))
# ── 2. 台词本:全片台词合计 ──
with open(dialogue_path, encoding='utf-8') as f:
dialogue = f.read()
dtm = re.search(r'全片台词合计约\s*([\d.]+)\s*秒', dialogue)
dialogue_total = float(dtm.group(1)) if dtm else None
if dialogue_total is None:
# 退回:汇总每个场景的 **合计** / 约Xs秒
scene_sum = re.findall(r'\*\*约\s*([\d.]+)\s*秒\*\*', dialogue)
if scene_sum:
dialogue_total = round(sum(float(x) for x in scene_sum), 1)
print("=" * 76)
print("导演讲戏本预算自检报告")
print("=" * 76)
print(f"讲戏本: {director_path}")
print(f"台词本: {dialogue_path}")
print()
budget_sum = round(sum(b for _, b in p_budgets), 1)
print(f"讲戏本剧情点: {len(p_budgets)} 个({', '.join(p for p, _ in p_budgets) if p_budgets else '未解析到'})")
print(f"各 P 预算之和: {budget_sum} 秒")
if header_total is not None:
flag = "(一致)" if abs(budget_sum - header_total) < 0.5 else "(⚠ 与头部声明不一致)"
print(f"头部声明总预算: {header_total} 秒 {flag}")
print(f"台词本全片台词合计: {dialogue_total if dialogue_total is not None else '未解析到'} 秒")
print()
hard_errors = 0
soft_warnings = 0
# 硬错误 1:预算略小于台词
if dialogue_total is not None and budget_sum + 1e-9 < dialogue_total:
diff = round(dialogue_total - budget_sum, 1)
print(f"✗ [硬错误] 各 P 预算之和 {budget_sum}s < 台词合计 {dialogue_total}s,差 {diff}s:台词塞不下,动作/反应余量为负,分镜阶段必然返工。")
print(f" 请为相关 P 增加动作/反应预算('时长'字段应按 台词参考时长 + 视觉动作预算 填写)。")
hard_errors += 1
elif dialogue_total is not None:
margin = round(budget_sum - dialogue_total, 1)
print(f"✓ 预算之和 {budget_sum}s >= 台词合计 {dialogue_total}s,动作/反应余量 {margin}s")
if margin > HIGH_MARGIN:
print(f" ⚠ [软提示] 余量 {margin}s 偏高(>{HIGH_MARGIN}s),可能为凑时长加戏,请人工确认是否合理。")
soft_warnings += 1
if margin < 0.5:
print(f" ⚠ [软提示] 余量仅 {margin}s,几乎无动作/反应空间;若含无台词表演点可能偏紧。")
soft_warnings += 1
# 硬错误 2:单 P 超 15s
over = [(p, b) for p, b in p_budgets if b > MOTHER_MAX]
if over:
print(f"✗ [硬错误] 以下剧情点预算超过母镜头 {MOTHER_MAX}s 硬上限:{', '.join(f'{p}={b}s' for p, b in over)}")
print(" 需拆分或重新分配,否则分镜母镜头超 15s。")
hard_errors += 1
print()
print("=" * 76)
if hard_errors == 0 and soft_warnings == 0:
print("✓ 讲戏本预算自洽,可进入下一阶段。")
sys.exit(0)
elif hard_errors == 0:
print(f"✓ 预算自洽({soft_warnings} 个软提示,请人工参考)。")
sys.exit(0)
else:
print(f"✗ 发现 {hard_errors} 个硬错误(必须修复)。")
sys.exit(1)
if __name__ == '__main__':
main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 198k | Updated | Claude CodeCodex | Verified | Install |
| 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: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| 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: 194k | Updated | Claude CodeCodex | Verified | Install |