Skip to content

joeseesun/qiaomu-cut-skill

qiaomu-cut

Verified

把一句话需求转成可复现、可验收视频工程的乔木智能剪辑导演。Use when the user asks to create, plan, edit, remix, explain, narrate, subtitle, animate, composite, or render a video—including one-line requests such as “制作一个科普视频:介绍 LLM 中的 RL”. Builds English-learning movie mixes, person profiles, explainers, code-drawn motion graphics (UI morph loops, showreels, kinetic-type promos, line-art/pixel/Three.js animation, lyric MVs), product launch videos, cinematic shorts, social clips, course/PPT videos, AI image/video/TTS/music-assisted stories, stock-footage stories, or multi-source montage. Routes across 33台词, ClipSeek/open media, local files, ListenHub, image generation, HTML/SVG/Canvas motion graphics with beat-grid storyboards and determinism probes, Manim, ffmpeg-full, captions, sound design, frame-0 video covers with YouTube thumbnails, and release checks.

FileKindSize
SKILL.mdskill md54.0 KB
agents/interface.yamlconfig17.2 KB
agents/openai.yamlconfig323 B
assets/explainer-social/base.cssother24.9 KB
assets/explainer-social/component-runtime.jsscript41.1 KB
assets/explainer-social/scene-shell.htmlother829 B
assets/motion-kit/qiaocut-motion.jsscript8.2 KB
assets/qiaomu-profile/qiaomu_avatar.jpegbinary5.6 MB
assets/qiaomu-profile/qiaomu_reward_qr.pngbinary515.9 KB
assets/qiaomu-profile/qiaomu_wechat_public_account_qr.jpgbinary8.3 KB
references/ad-concept-library.mdmarkdown6.5 KB
references/aesthetic-preferences.mdmarkdown3.0 KB
references/brand-card-templates.mdmarkdown1.6 KB
references/chinese-font-themes.mdmarkdown804 B
references/cinematic-techniques.mdmarkdown2.8 KB
references/code-motion-direction.mdmarkdown11.1 KB
references/cover-thumbnail-method.mdmarkdown6.4 KB
references/english-mix-fast-path.mdmarkdown6.0 KB
references/explainer-autopilot.mdmarkdown11.1 KB
references/feedback-evolution.mdmarkdown37.6 KB
references/ffmpeg-full.mdmarkdown1.7 KB
references/incremental-production.mdmarkdown4.4 KB
references/licensing.mdmarkdown1.9 KB
references/listenhub-provider.mdmarkdown12.6 KB
references/motion-prompt-patterns.mdmarkdown9.3 KB
references/movie-montage-fast-path.mdmarkdown4.4 KB
references/qiaocut-ir.mdmarkdown4.2 KB
references/renderer-engines.mdmarkdown3.3 KB
references/schemas/qiaocut.timeline.v1.schema.jsonconfig7.2 KB
references/social-safe-zones.mdmarkdown3.3 KB
references/source-adapters.mdmarkdown5.7 KB
references/source-selection-gate.mdmarkdown2.5 KB
references/styles/cinematic.mdmarkdown475 B
references/styles/explainer.mdmarkdown1.5 KB
references/styles/social-short.mdmarkdown383 B
references/timeline-schema.mdmarkdown12.2 KB
references/trust-boundary.mdmarkdown3.0 KB
references/workflows.mdmarkdown7.5 KB
scripts/adapters/33tc.jsscript2.1 KB
scripts/adapters/33tc_cli.jsscript4.8 KB
scripts/adapters/clipseek.jsexecutablescript5.4 KB
scripts/adapters/listenhub.jsscript35.7 KB
scripts/adapters/listenhub_cli.jsexecutablescript2.1 KB
scripts/adapters/local.jsscript2.4 KB
scripts/audio_beats.jsscript17.5 KB
scripts/bilingual_typography_smoke.jsscript1.4 KB
scripts/bootstrap_listenhub.shexecutablescript5.4 KB
scripts/bootstrap_macos.shexecutablescript2.5 KB
scripts/brand_templates.jsscript2.4 KB
scripts/brand_templates_smoke.jsscript869 B
scripts/complete_sentence.jsscript3.9 KB
scripts/component_motion_smoke.jsscript21.2 KB
scripts/concept_smoke.jsscript5.6 KB
scripts/cover.jsscript9.1 KB
scripts/cover_smoke.jsscript5.1 KB
scripts/english_mix.jsscript24.1 KB
scripts/english_mix_smoke.jsscript5.7 KB
scripts/explainer_beats.jsscript6.7 KB
scripts/explainer_beats_smoke.jsscript4.8 KB
scripts/explainer_cognitive_timing_smoke.jsscript5.8 KB
scripts/explainer_e2e_smoke.jsscript7.4 KB
scripts/explainer_pipeline.jsscript49.3 KB
scripts/explainer_pipeline_smoke.jsscript12.7 KB
scripts/explainer_spec.jsscript51.4 KB
scripts/explainer_timing.jsscript8.6 KB
scripts/fetch_generated.jsexecutablescript17.4 KB
scripts/ffmpeg_retry_shim_smoke.jsscript2.3 KB
scripts/file_narration_smoke.jsexecutablescript6.6 KB
scripts/ingest_asset.jsexecutablescript15.0 KB
scripts/kinetic_caption_smoke.jsscript1.4 KB
scripts/license_report.jsexecutablescript2.3 KB
scripts/listenhub_narration.jsexecutablescript11.0 KB
scripts/listenhub_smoke.jsexecutablescript26.7 KB
scripts/local_preferences.jsscript1.8 KB
scripts/motion_studio.jsscript55.9 KB
scripts/motion_studio_smoke.jsscript11.0 KB
scripts/movie_montage.jsscript24.7 KB
scripts/movie_montage_smoke.jsscript4.0 KB
scripts/qcut.jsexecutablescript40.8 KB
scripts/release_check.jsscript7.0 KB
scripts/render_project.jsscript85.1 KB
scripts/render_scene.jsscript26.4 KB
scripts/renderers/ass.jsexecutablescript2.7 KB
scripts/renderers/bilingual_ass.jsscript12.6 KB
scripts/renderers/html_scene.jsexecutablescript3.9 KB
scripts/renderers/procedural_music.jsscript7.6 KB
scripts/scene_content_cache.jsscript3.2 KB
scripts/scene_renderer_smoke.jsscript2.8 KB
scripts/security_smoke.jsscript10.2 KB
scripts/sentence_boundary_smoke.jsscript1.6 KB
scripts/setup_deps.jsscript8.4 KB
scripts/shims/ffmpegexecutablescript3.0 KB
scripts/source_review.jsexecutablescript5.1 KB
scripts/timeline_features_smoke.jsscript3.7 KB
scripts/trigger_eval.pyexecutablescript2.5 KB
scripts/validate_skill.pyexecutablescript3.1 KB
scripts/verify_marswave_vendor.jsexecutablescript3.5 KB
scripts/verify_video.jsexecutablescript3.4 KB

scripts/trigger_eval.py

View on GitHub
#!/usr/bin/env python3
import argparse
import json
from pathlib import Path


TRIGGER_HINTS = [
    "视频", "剪辑", "混剪", "字幕", "电影", "素材", "片头", "转场",
    "运镜", "口播", "科普", "介绍", "short", "video", "edit", "montage",
    "动态图形", "动效视频", "动画", "showreel", "mv", "motion graphic", "kinetic typography"
]

EXPLICIT_NON_VIDEO_HINTS = [
    "不需要视频", "不要视频", "仅音频", "只要音频", "audio only", "no video"
]


def should_trigger(text: str) -> bool:
    lowered = text.lower()
    if any(hint.lower() in lowered for hint in EXPLICIT_NON_VIDEO_HINTS):
        return False
    return any(hint.lower() in lowered for hint in TRIGGER_HINTS)


def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("skill_dir")
    parser.add_argument("--cases", default="evals/trigger_cases.json")
    parser.add_argument("--output", default="reports/trigger-eval.json")
    args = parser.parse_args()

    root = Path(args.skill_dir).resolve()
    cases_path = root / args.cases
    cases = json.loads(cases_path.read_text(encoding="utf-8"))
    results = []
    failures = []

    for case in cases.get("should_trigger", []):
        got = should_trigger(case["input"])
        results.append({"id": case["id"], "expected": True, "got": got})
        if not got:
            failures.append(case["id"])

    for case in cases.get("should_not_trigger", []) + cases.get("near_neighbor", []):
        got = should_trigger(case["input"])
        results.append({"id": case["id"], "expected": False, "got": got})
        if got:
            failures.append(case["id"])

    report = {
        "ok": not failures,
        "total": len(results),
        "summary": {
            "total": len(results),
            "passed": len(results) - len(failures),
            "false_positive": sum(1 for r in results if r["got"] and not r["expected"]),
            "false_negative": sum(1 for r in results if r["expected"] and not r["got"]),
            "pass_rate": round((len(results) - len(failures)) / len(results), 3) if results else 0.0,
        },
        "failures": failures,
        "results": results,
    }
    output = root / args.output
    output.parent.mkdir(parents=True, exist_ok=True)
    output.write_text(json.dumps(report, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
    print(json.dumps(report, ensure_ascii=False, indent=2))
    return 0 if not failures else 1


if __name__ == "__main__":
    raise SystemExit(main())
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 322kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall