PhilRobinluo/ai-coevolution-skills / skills/lls-video-digester
lls-video-digester
Verified把用户提供或公开可用的字幕整理成带时间证据的观点、质疑点和行动清单。 当用户要处理相关材料并需要可验证交付物时使用。
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 1.8 KB |
| SKILL.md | skill md | 4.8 KB |
| references/evidence-note-template.md | markdown | 429 B |
| scripts/index_transcript.pyexecutable | script | 2.3 KB |
| tests/test_index_transcript.py | script | 715 B |
scripts/index_transcript.py
View on GitHub#!/usr/bin/env python3
"""把用户合法取得的 SRT/VTT 字幕转成可引用的时间证据索引。"""
from __future__ import annotations
import argparse,json,re
from pathlib import Path
TIME=re.compile(r'(?:(\d+):)?(\d{2}):(\d{2})[,.](\d{3})')
def sec(s):
m=TIME.search(s); return None if not m else int(m.group(1) or 0)*3600+int(m.group(2))*60+int(m.group(3))+int(m.group(4))/1000
def fmt(v): return f'{int(v)//3600:02d}:{int(v)%3600//60:02d}:{int(v)%60:02d}'
def parse(text):
blocks=re.split(r'\n\s*\n',text.replace('\r','')); rows=[]
for b in blocks:
lines=[x.strip() for x in b.splitlines() if x.strip() and x.strip()!='WEBVTT'];
if lines and lines[0].isdigit(): lines=lines[1:]
if not lines: continue
idx=next((i for i,x in enumerate(lines) if '-->' in x),None)
if idx is None: continue
a,btime=lines[idx].split('-->',1); start=sec(a); end=sec(btime); content=' '.join(lines[idx+1:]); content=re.sub(r'<[^>]+>','',content)
if start is not None and end is not None and content: rows.append({'start':start,'end':end,'text':content})
return rows
def main():
p=argparse.ArgumentParser(); p.add_argument('input',type=Path); p.add_argument('output',type=Path); p.add_argument('--chunk-seconds',type=int,default=120); a=p.parse_args()
if not a.input.is_file(): raise SystemExit('字幕文件不存在')
if not 30<=a.chunk_seconds<=900: raise SystemExit('--chunk-seconds 必须在 30 到 900')
rows=parse(a.input.read_text(errors='replace'));
if not rows: raise SystemExit('未解析到带时间字幕')
chunks=[]; current=[]; start=rows[0]['start']
for r in rows:
if current and r['start']-start>=a.chunk_seconds: chunks.append({'start':start,'end':current[-1]['end'],'text':' '.join(x['text'] for x in current)}); current=[]; start=r['start']
current.append(r)
if current: chunks.append({'start':start,'end':current[-1]['end'],'text':' '.join(x['text'] for x in current)})
out={'source':str(a.input),'segments':len(rows),'chunks':[{'start':fmt(x['start']),'end':fmt(x['end']),'text':x['text']} for x in chunks]}
a.output.parent.mkdir(parents=True,exist_ok=True); a.output.write_text(json.dumps(out,ensure_ascii=False,indent=2)+'\n'); print(json.dumps({'segments':len(rows),'chunks':len(chunks),'output':str(a.output)},ensure_ascii=False))
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 |