agentleFS
full-aigc-skillsfull-aigc-skills/minimax-skills/skills/minimax-multimodal-toolkit/SKILL.md

minimax-multimodal-toolkit

Skill4 starsChanged 57 days agoApache-2.0View on GitHub

Use mmx to generate text, images, video, speech, and music via the MiniMax AI platform. Use when the user wants to create media content, chat with MiniMax models, perform web search, or manage MiniMax API resources from the terminal.

  • Reads credentials
  • Installs packages

7 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use mmx to generate text, images, video, speech, music, and perform web search via the MiniMax AI platform.

    line 8
  2. Distinctive:

    Always use these flags in non-interactive (agent/CI) contexts:

    line 29
  3. Distinctive:

    Use this to dynamically register mmx commands as tools in your agent framework.

    line 308

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

MiniMax CLI — Agent Skill Guide

Use mmx to generate text, images, video, speech, music, and perform web search via the MiniMax AI platform.

Prerequisites

# Install
npm install -g mmx-cli

# Auth (persisted to ~/.mmx/credentials.json)
mmx auth login --api-key sk-xxxxx

# Or pass per-call
mmx text chat --api-key sk-xxxxx --message "Hello"

Region is auto-detected. Override with --region global or --region cn.


Agent Flags

Always use these flags in non-interactive (agent/CI) contexts:

FlagPurpose
--non-interactiveFail fast on missing args instead of prompting
--quietSuppress spinners/progress; stdout is pure data
--output jsonMachine-readable JSON output
--asyncReturn task ID immediately (video generation)
--dry-runPreview the API request without executing
--yesSkip confirmation prompts

Commands

text chat

Chat completion. Default model: MiniMax-M2.7.

mmx text chat --message <text> [flags]
FlagTypeDescription
--message <text>string, required, repeatableMessage text. Prefix with role: to set role (e.g. "system:You are helpful", "user:Hello")
--messages-file <path>stringJSON file with messages array. Use - for stdin
--system <text>stringSystem prompt
--model <model>stringModel ID (default: MiniMax-M2.7)
--max-tokens <n>numberMax tokens (default: 4096)
--temperature <n>numberSampling temperature (0.0, 1.0]
--top-p <n>numberNucleus sampling threshold
--streambooleanStream tokens (default: on in TTY)
--tool <json-or-path>string, repeatableTool definition JSON or file path
# Single message
mmx text chat --message "user:What is MiniMax?" --output json --quiet

# Multi-turn
mmx text chat \
  --system "You are a coding assistant." \
  --message "user:Write fizzbuzz in Python" \
  --output json

# From file
cat conversation.json | mmx text chat --messages-file - --output json

stdout: response text (text mode) or full response object (json mode).


image generate

Generate images. Model: image-01.

mmx image generate --prompt <text> [flags]
FlagTypeDescription
--prompt <text>string, requiredImage description
--aspect-ratio <ratio>stringe.g. 16:9, 1:1
--n <count>numberNumber of images (default: 1)
--subject-ref <params>stringSubject reference: type=character,image=path-or-url
--out-dir <dir>stringDownload images to directory
--out-prefix <prefix>stringFilename prefix (default: image)
mmx image generate --prompt "A cat in a spacesuit" --output json --quiet
# stdout: image URLs (one per line in quiet mode)

mmx image generate --prompt "Logo" --n 3 --out-dir ./gen/ --quiet
# stdout: saved file paths (one per line)

video generate

Generate video. Default model: MiniMax-Hailuo-2.3. This is an async task — by default it polls until completion.

mmx video generate --prompt <text> [flags]
FlagTypeDescription
--prompt <text>string, requiredVideo description
--model <model>stringMiniMax-Hailuo-2.3 (default) or MiniMax-Hailuo-2.3-Fast
--first-frame <path-or-url>stringFirst frame image
--callback-url <url>stringWebhook URL for completion
--download <path>stringSave video to specific file
--asyncbooleanReturn task ID immediately
--no-waitbooleanSame as --async
--poll-interval <seconds>numberPolling interval (default: 5)
# Non-blocking: get task ID
mmx video generate --prompt "A robot." --async --quiet
# stdout: {"taskId":"..."}

# Blocking: wait and get file path
mmx video generate --prompt "Ocean waves." --download ocean.mp4 --quiet
# stdout: ocean.mp4

video task get

Query status of a video generation task.

mmx video task get --task-id <id> [--output json]

video download

Download a completed video by task ID.

mmx video download --file-id <id> [--out <path>]

speech synthesize

Text-to-speech. Default model: speech-2.8-hd. Max 10k chars.

mmx speech synthesize --text <text> [flags]
FlagTypeDescription
--text <text>stringText to synthesize
--text-file <path>stringRead text from file. Use - for stdin
--model <model>stringspeech-2.8-hd (default), speech-2.6, speech-02
--voice <id>stringVoice ID (default: English_expressive_narrator)
--speed <n>numberSpeed multiplier
--volume <n>numberVolume level
--pitch <n>numberPitch adjustment
--format <fmt>stringAudio format (default: mp3)
--sample-rate <hz>numberSample rate (default: 32000)
--bitrate <bps>numberBitrate (default: 128000)
--channels <n>numberAudio channels (default: 1)
--language <code>stringLanguage boost
--subtitlesbooleanInclude subtitle timing data
--pronunciation <from/to>string, repeatableCustom pronunciation
--sound-effect <effect>stringAdd sound effect
--out <path>stringSave audio to file
--streambooleanStream raw audio to stdout
mmx speech synthesize --text "Hello world" --out hello.mp3 --quiet
# stdout: hello.mp3

echo "Breaking news." | mmx speech synthesize --text-file - --out news.mp3

music generate

Generate music. Model: music-2.5. Responds well to rich, structured descriptions.

mmx music generate --prompt <text> [--lyrics <text>] [flags]
FlagTypeDescription
--prompt <text>stringMusic style description (can be detailed)
--lyrics <text>stringSong lyrics with structure tags. Use "\u65e0\u6b4c\u8bcd" for instrumental. Cannot be used with --instrumental
--lyrics-file <path>stringRead lyrics from file. Use - for stdin
--vocals <text>stringVocal style, e.g. "warm male baritone", "bright female soprano", "duet with harmonies"
--genre <text>stringMusic genre, e.g. folk, pop, jazz
--mood <text>stringMood or emotion, e.g. warm, melancholic, uplifting
--instruments <text>stringInstruments to feature, e.g. "acoustic guitar, piano"
--tempo <text>stringTempo description, e.g. fast, slow, moderate
--bpm <number>numberExact tempo in beats per minute
--key <text>stringMusical key, e.g. C major, A minor, G sharp
--avoid <text>stringElements to avoid in the generated music
--use-case <text>stringUse case context, e.g. "background music for video", "theme song"
--structure <text>stringSong structure, e.g. "verse-chorus-verse-bridge-chorus"
--references <text>stringReference tracks or artists, e.g. "similar to Ed Sheeran"
--extra <text>stringAdditional fine-grained requirements
--instrumentalbooleanGenerate instrumental music (no vocals). Cannot be used with --lyrics or --lyrics-file
--aigc-watermarkbooleanEmbed AI-generated content watermark
--format <fmt>stringAudio format (default: mp3)
--sample-rate <hz>numberSample rate (default: 44100)
--bitrate <bps>numberBitrate (default: 256000)
--out <path>stringSave audio to file
--streambooleanStream raw audio to stdout

At least one of --prompt or --lyrics is required.

# Simple usage
mmx music generate --prompt "Upbeat pop" --lyrics "La la la..." --out song.mp3 --quiet

# Detailed prompt with vocal characteristics
mmx music generate --prompt "Warm morning folk" \
  --vocals "male and female duet, harmonies in chorus" \
  --instruments "acoustic guitar, piano" \
  --bpm 95 \
  --lyrics-file song.txt \
  --out duet.mp3

# Instrumental (use --instrumental flag)
mmx music generate --prompt "Cinematic orchestral, building tension" --instrumental --out bgm.mp3

vision describe

Image understanding via VLM. Provide either --image or --file-id, not both.

mmx vision describe (--image <path-or-url> | --file-id <id>) [flags]
FlagTypeDescription
--image <path-or-url>stringLocal path or URL (auto base64-encoded)
--file-id <id>stringPre-uploaded file ID (skips base64)
--prompt <text>stringQuestion about the image (default: "Describe the image.")
mmx vision describe --image photo.jpg --prompt "What breed?" --output json

stdout: description text (text mode) or full response (json mode).


search query

Web search via MiniMax.

mmx search query --q <query>
FlagTypeDescription
--q <query>string, requiredSearch query
mmx search query --q "MiniMax AI" --output json --quiet

quota show

Display Token Plan usage and remaining quotas.

mmx quota show [--output json]

Tool Schema Export

Export all commands as Anthropic/OpenAI-compatible JSON tool schemas:

# All tool-worthy commands (excludes auth/config/update)
mmx config export-schema

# Single command
mmx config export-schema --command "video generate"

Use this to dynamically register mmx commands as tools in your agent framework.


Exit Codes

CodeMeaning
0Success
1General error
2Usage error (bad flags, missing args)
3Authentication error
4Quota exceeded
5Timeout
10Content filter triggered

Piping Patterns

# stdout is always clean data — safe to pipe
mmx text chat --message "Hi" --output json | jq '.content'

# stderr has progress/spinners — discard if needed
mmx video generate --prompt "Waves" 2>/dev/null

# Chain: generate image → describe it
URL=$(mmx image generate --prompt "A sunset" --quiet)
mmx vision describe --image "$URL" --quiet

# Async video workflow
TASK=$(mmx video generate --prompt "A robot" --async --quiet | jq -r '.taskId')
mmx video task get --task-id "$TASK" --output json
mmx video download --task-id "$TASK" --out robot.mp4

Configuration Precedence

CLI flags → environment variables → ~/.mmx/config.json → defaults.

# Persistent config
mmx config set --key region --value cn
mmx config show

# Environment
export MINIMAX_API_KEY=sk-xxxxx
export MINIMAX_REGION=cn
<!-- QUALITY_BASELINE_V1 -->

When to use(什么时候使用)

当用户需要 为当前请求选择并执行可验证、可恢复的专业工作流 时加载本技能。先从请求中提取目标、输入、约束、交付格式和验收标准;描述摘要为:Use mmx to generate text, images, video, speech, and music via the MiniMax AI platform. Use when the user wants to create media content, chat with MiniMax models, perform web search, or manage MiniMax API resources from the terminal.。

Rules

  • 先读后写:先确认当前状态与真实能力,再执行会改变外部状态的动作。
  • 权限最小化:只使用完成当前步骤所需的文件、工具、账户与网络范围。
  • 证据优先:运行结果、资源 ID、版本、哈希或测试输出缺失时,明确标记为 NOT_VERIFIED。
  • 幂等优先:保留请求标识与阶段状态;结果不明确时先查询,不进行盲目重试。
  • 隐私安全:日志、示例、回执和错误信息不得包含 token、cookie、密钥或个人敏感数据。

Workflow

Step 1:澄清意图

确认本技能是否匹配目标;若只是相邻需求,交给更精确的技能。

Step 2:执行预检

确认目标、输入、约束、可用工具、成功标准和失败边界;任一关键条件未知时停止在只读阶段。

Step 3:形成计划

列出将调用的工具、会改变的对象、成功标准以及失败后的安全退出方式。

Step 4:执行动作

按最小充分步骤执行,并在关键状态变化处记录证据;每个外部调用均保留可关联的状态或回执。

Step 5:验证交付

输出结果、验证证据、未完成项、风险和明确的下一步,并把事实、推断和未验证项分开陈述。

Validation checklist

  • 技能触发条件与用户意图一致,没有把相邻任务误路由到本技能。
  • 输入、目标对象、版本和输出位置均已明确,且没有使用猜测值替代必填值。
  • 所有写入、付费、发布或不可逆动作都在用户授权范围内。
  • 结果已用独立检查验证;仅有“命令成功”或“文件存在”不算完整验收。
  • 输出包含实际证据、失败/跳过项、剩余风险和可执行的下一步。

Gotchas

  1. 把计划当结果:文档或提示词不等于真实执行;必须标明实际运行层级。
  2. 错误重试:超时或响应丢失可能已经产生远端状态,先查询再决定是否重试。
  3. 隐式扩大范围:批量、全量、发布、覆盖和付费不是普通读写的自然延伸。
  4. 版本漂移:引用外部资源时记录版本、tag 或提交;不要把可变分支当发布证据。
  5. 证据过期:缓存、旧截图和历史测试不能证明当前环境;在交付前刷新关键证据。

不适用与边界

不超出用户给定范围;写入、付费、发布和不可逆动作需要明确授权。 如果请求需要别的技能,不复制其正文;按技能名称进行交接,并保留当前任务上下文。

Progressive disclosure

  • 需要确定输入/输出、状态和授权点时,读取 references/workflow-contract.md。
  • 需要交付前自检时,读取 references/validation-checklist.md。
  • 遇到超时、部分成功或恢复场景时,读取 references/error-recovery.md。
  • 首次运行、拒绝越权和失败恢复分别参考 examples/happy-path.md、examples/boundary-refusal.md、examples/failure-recovery.md。

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.