minimax-av SKILL.md skillA
minimax-av from noah-1106/skills_repo is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add noah-1106/skills_repo/minimax-av@git:20260908.5ebc0caPins git:20260908.5ebc0ca in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- noah-1106/skills_repo
- Path
skills/minimax-av/SKILL.md- Versions recorded
- 4
- Latest version
- git:20260908.5ebc0ca, committed Sep 8, 2026
- First version recorded
- Sep 7, 2026
- Size
- 5,118 bytes
- Repository stars
- 32
- License
- MIT
MiniMax 全模态理解 CLI，双子命令。transcribe：音频转文本（asr-1.0，8 种格式 ≤50MB/≤500s，支持 SRT/VTT 字幕与说话人分离时间戳）；understand：视觉理解（MiniMax-M3 原生，视频 mp4/mov 等 + 图片 png/jpg/webp/gif，≤100MB 可混合多文件，问答/摘要/JSON 输出）。模型、端点、API Key 全部走 config.json 可换。用于录音转写、字幕生产、视频分析、图片理解、会议纪要。.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# minimax-av — MiniMax 全模态理解（音频转写 + 视觉理解） ## 场景 - **transcribe**：录音/语音备忘录/访谈转文字（多格式、≤500 秒、可出 SRT/VTT 字幕与说话人分离时间戳） - **understand**：视频内容问答/摘要（M3 原生视频理解）+ 图片理解/对比/文字提取（可混合多文件） ## 触发条件 用户给出音频文件要求转写/出字幕，或给出视频/图片文件要求理解、描述、分析、问答时。 ## 输入（硬限制） | 子命令 | 文件 | 限制 | |---|---|---| | `transcribe` | wav/aiff/flac/m4a/mp3/aac/opus/ogg | ≤**50MB**（413）、≤**500 秒**（400，超长不截断直接拒）；裸 PCM 不支持 | | `understand` | 视频 mp4/mov/avi/mkv/webm/flv/wmv/mpeg；**图片 png/jpg/webp/gif** | 单文件 ≤100MB（config 可调）；**图片 >2MB 建议先缩放**（脚本会提示）；时长未标注 | 参数：`--out 文件`（落盘）、`--show-usage`（stderr 打印延迟/用量）；transcribe 另有 **`--format`**；understand 另有 `--prompt`、`--json`、`--max-tokens`。 ### 时间戳/字幕（transcribe 的 --format） | format | 输出 | 适合 | |---|---|---| | `json`（默认） | 纯文本 + duration | 快速转写 | | `verbose_json` | **句级时间戳 + 说话人分离**（[S1] 0.02s-5.32s 文本，含 n_speakers） | 会议纪要、访谈 | | `srt` / `vtt` | **标准字幕格式**（00:00:00,020 --> 00:00:05,320） | 视频配字幕 | 注意：这三种格式启用时间戳对齐与说话人分离，不能与流式同用（本 skill 非流式，无此限制）。 ## 步骤 ```bash python3 scripts/minimax_av.py transcribe 录音.mp3 python3 scripts/minimax_av.py transcribe meeting.mp3 --format srt --out 字幕.srt python3 scripts/minimax_av.py transcribe interview.wav --format verbose_json # 说话人+时间戳 …
Read the whole file at its exact version.
How to install
mdr add noah-1106/skills_repo/minimax-av@sha256:18389e72f75a712eThe line at the top of this page pins the label git:20260908.5ebc0ca, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_44fo73eo2kfuzvbw)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Sep 7, 2026 and Sep 8, 2026, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260908.5ebc0ca latest | 2026-09-08 | 5ebc0ca | 5,118 B | A | view · diff |
| git:20260908.d9249a8 | 2026-09-08 | d9249a8 | 5,642 B | A | view · diff |
| git:20260907.77a1cf3 | 2026-09-07 | 77a1cf3 | 4,694 B | A | view · diff |
| git:20260907.32b6c64 | 2026-09-07 | 32b6c64 | 3,369 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (5118 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
noah-1106/skills_repo · 32 stars · license MIT · pushed 2026-11-02 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_44fo73eo2kfuzvbw GET https://markdownregistry.com/api/v1/resolve?ref=noah-1106/skills_repo/minimax-av GET https://markdownregistry.com/api/v1/blob/18389e72f75a712e1fa25779a25bbbe7c01b941300e0a7f4ef18664c5eccce48
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.