MiniMax-AI/mcode-stack / stacks/jiachen/skills/making-content
making-content
VerifiedContent production in three modes: rewrite, Xiaohongshu (小红书) image post, short video. Rewrite (default): turns an AI-sounding Chinese or mixed Chinese-English draft (post, newsletter, engineering blog, commentary) into the author's restrained style: few emoji, short paragraphs, hedged claims, neutral wording, English terms split by spaces. Triggers: 按我的风格改, 去 AI 味, 润色, 改改这篇, or a vague 帮我改改 when the context is a structured draft. Also for a Xiaohongshu image post from scratch (title, cover, cards, body, hooks); read references/xhs-post.md first. Also turns a document, research, notes or an article into a 60-180 s vertical video (抖音, 小红书, 视频号, B站: real sources, voice-over, captions, cover). Triggers: 做成视频, 做个短视频, 剪个 90 秒版本, 配音, 换成我的声音, or complaints of 字太密, AI 味太重, 容易走神; read references/video-pipeline.md first.
| File | Kind | Size |
|---|---|---|
| .gitignore | other | 167 B |
| SKILL.md | skill md | 25.6 KB |
| evals/evals.json | config | 4.5 KB |
| references/layouts.md | markdown | 8.5 KB |
| references/text-density-lint.md | markdown | 4.2 KB |
| references/video-pipeline.md | markdown | 12.5 KB |
| references/voice-and-music.md | markdown | 5.5 KB |
| references/xhs-post.md | markdown | 9.0 KB |
| requirements.txt | manifest | 233 B |
| scripts/capture.py | script | 6.4 KB |
| scripts/compositor.html | other | 27.8 KB |
| scripts/fetch_assets.py | script | 4.0 KB |
| scripts/lint.py | script | 8.6 KB |
| scripts/render.py | script | 4.6 KB |
| scripts/timeline.py | script | 8.7 KB |
| scripts/tts.py | script | 5.5 KB |
| templates/cover.html | other | 2.9 KB |
| templates/script.example.json | config | 5.0 KB |
| templates/sources.example.json | config | 58 B |
evals/evals.json
View on GitHub{
"skill_name": "making-content",
"note": "expected_behavior describes a good run of the whole pipeline or of one revision step, not a single file. Prompts are in Chinese, as the user writes them.",
"evals": [
{
"id": 1,
"name": "doc-to-douyin-first-draft",
"prompt": "把这篇调研笔记做成一个短视频,我想发抖音。需要我提供什么 api key 吗?",
"expected_behavior": [
"Asks one AskUserQuestion round before building: platform and length, voice, hook options drawn from the note, music, and what personal or employer information to keep out",
"Says in that first message what only the user can do (Volcengine voice clone, speaker id and API key) and that work proceeds with the free Edge voice meanwhile",
"Removes the private details the user names from the narration, captions and every on-screen text before anything is rendered",
"Writes script.json with a hook in line 0 and meta.hook, a 先说结论 line within about 15 s, spoken-style narration of roughly 430-450 字 for 90 s, and cap text with digits and '/' breaks",
"Rotates layouts with a new shot every 3-6 s, repeating a layout back to back only for a pair; English quotes carry a larger Chinese translation",
"Runs render.py --stills, looks at sheet.jpg with the Read tool, and loops lint.py until it prints 0 issue(s)",
"Renders a 1080x1920 mp4 under 30 MB at about -14 LUFS, checks it with ffprobe and ebur128, and sends it with the cover",
"Gives a description of about 400 字 with the conclusion, the sources and the music credit line, and adds no 下期预告 ending"
],
"files": []
},
{
"id": 2,
"name": "switch-to-cloned-voice",
"prompt": "我在火山上创建了一个音色 S_Example123,还有一个临时的 api key 00000000-1111-2222-3333-444444444444,换成我的声音试试,先给我听个样本",
"expected_behavior": [
"Passes the key and speaker id through VOLC_TTS_API_KEY and VOLC_TTS_SPEAKER only; the key appears in no file, script or commit",
"Voices one representative line in 2-3 variants (speech_rate and the style instruction) and sends them for the user to choose before voicing the whole script",
"On 45000030 tells the user to enable 声音复刻 2.0 字符版 and grant the key; on 45000292 switches to the seed-icl-2.0 字符版 resource instead of retrying",
"Writes the cloned voice into a new --out directory so the approved Edge or earlier clips are not overwritten",
"Checks that identifiers such as rusty_v8 are not read as 杠, then rebuilds the timeline and mix with the new clips",
"Suggests revoking the temporary key once the video is final"
],
"files": []
},
{
"id": 3,
"name": "too-dense-revision",
"prompt": "还是字体太密,有点眼花,字幕可以放得下面一些吗?多一些留白",
"expected_behavior": [
"Runs lint.py on the current timeline and reports which shots and captions fail and why before changing anything",
"Fixes failures with the text-density-lint table: shorter text, a second shot, context dimmed to texture, '/' moved in cap, rather than shrinking fonts below 34 px",
"Keeps captions at one line of at most 16 字 with the top at y 1380 or lower, and keeps the keyword sticker in the top zone",
"Where a frame becomes empty, adds motion (stickers, counter, bars, ambient shapes) instead of more words",
"Re-renders stills and shows the contact sheet before spending time on a full render",
"Ends with lint at 0 issue(s) and a re-rendered video"
],
"files": []
},
{
"id": 4,
"name": "cut-a-90-second-version",
"prompt": "这个 3 分钟的视频发抖音太长了吧?你剪一个 90 秒的版本,另外把那个 biu 的音效去掉",
"expected_behavior": [
"Answers the length question directly: 抖音 favours a 75-95 s main cut, the 3 minute cut stays for 小红书 or B站",
"Writes a separate shortened script (about 430-450 字) that keeps the hook, the early conclusion and the strongest evidence, leaving the long version's files untouched",
"Voices only new or changed lines, into a new tts directory",
"Confirms the mix has no per-element pop sounds and only a soft whoosh on layout-type changes",
"Re-lints and renders the short cut, and reports its length, size and loudness"
],
"files": []
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |