minimax-multimodal-toolkit

Name: minimax-multimodal-toolkit
Author: nguyenphp

39stars12forks

How to install minimax-multimodal-toolkit?

nguyenphp/antigravity-marketing/minimax-multimodal-toolkit

$npx -y skills add nguyenphp/antigravity-marketing --skill minimax-multimodal-toolkit

Installs into the current project.

Summary

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.

Files · 1

View on GitHub

SKILL.md

1	# MiniMax CLI — Agent Skill Guide
2
3	Use `mmx` to generate text, images, video, speech, music, and perform web search via the MiniMax AI platform.
4
5	## Prerequisites
6
7	```bash
8	# Install
9	npm install -g mmx-cli
10
11	# Auth (persisted to ~/.mmx/credentials.json)
12	mmx auth login --api-key sk-xxxxx
13
14	# Or pass per-call
15	mmx text chat --api-key sk-xxxxx --message "Hello"
16	```
17
18	Region is auto-detected. Override with `--region global` or `--region cn`.
19
20	---
21
22	## Agent Flags
23
24	Always use these flags in non-interactive (agent/CI) contexts:
25
26	\| Flag \| Purpose \|
27	\|---\|---\|
28	\| `--non-interactive` \| Fail fast on missing args instead of prompting \|
29	\| `--quiet` \| Suppress spinners/progress; stdout is pure data \|
30	\| `--output json` \| Machine-readable JSON output \|
31	\| `--async` \| Return task ID immediately (video generation) \|
32	\| `--dry-run` \| Preview the API request without executing \|
33	\| `--yes` \| Skip confirmation prompts \|
34
35	---
36
37	## Commands
38
39	### text chat
40
41	Chat completion. Default model: `MiniMax-M2.7`.
42
43	```bash
44	mmx text chat --message <text> [flags]
45	```
46
47	\| Flag \| Type \| Description \|
48	\|---\|---\|---\|
49	\| `--message <text>` \| string, required, repeatable \| Message text. Prefix with `role:` to set role (e.g. `"system:You are helpful"`, `"user:Hello"`) \|
50	\| `--messages-file <path>` \| string \| JSON file with messages array. Use `-` for stdin \|
51	\| `--system <text>` \| string \| System prompt \|
52	\| `--model <model>` \| string \| Model ID (default: `MiniMax-M2.7`) \|
53	\| `--max-tokens <n>` \| number \| Max tokens (default: 4096) \|
54	\| `--temperature <n>` \| number \| Sampling temperature (0.0, 1.0] \|
55	\| `--top-p <n>` \| number \| Nucleus sampling threshold \|
56	\| `--stream` \| boolean \| Stream tokens (default: on in TTY) \|
57	\| `--tool <json-or-path>` \| string, repeatable \| Tool definition JSON or file path \|
58
59	```bash
60	# Single message
61	mmx text chat --message "user:What is MiniMax?" --output json --quiet
62
63	# Multi-turn
64	mmx text chat \
65	--system "You are a coding assistant." \
66	--message "user:Write fizzbuzz in Python" \
67	--output json
68
69	# From file
70	cat conversation.json \| mmx text chat --messages-file - --output json
71	```
72
73	stdout: response text (text mode) or full response object (json mode).
74
75	---
76
77	### image generate
78
79	Generate images. Model: `image-01`.
80
81	```bash
82	mmx image generate --prompt <text> [flags]
83	```
84
85	\| Flag \| Type \| Description \|
86	\|---\|---\|---\|
87	\| `--prompt <text>` \| string, required \| Image description \|
88	\| `--aspect-ratio <ratio>` \| string \| e.g. `16:9`, `1:1` \|
89	\| `--n <count>` \| number \| Number of images (default: 1) \|
90	\| `--subject-ref <params>` \| string \| Subject reference: `type=character,image=path-or-url` \|
91	\| `--out-dir <dir>` \| string \| Download images to directory \|
92	\| `--out-prefix <prefix>` \| string \| Filename prefix (default: `image`) \|
93
94	```bash
95	mmx image generate --prompt "A cat in a spacesuit" --output json --quiet
96	# stdout: image URLs (one per line in quiet mode)
97
98	mmx image generate --prompt "Logo" --n 3 --out-dir ./gen/ --quiet
99	# stdout: saved file paths (one per line)
100	```
101
102	---
103
104	### video generate
105
106	Generate video. Default model: `MiniMax-Hailuo-2.3`. This is an async task — by default it polls until completion.
107
108	```bash
109	mmx video generate --prompt <text> [flags]
110	```
111
112	\| Flag \| Type \| Description \|
113	\|---\|---\|---\|
114	\| `--prompt <text>` \| string, required \| Video description \|
115	\| `--model <model>` \| string \| `MiniMax-Hailuo-2.3` (default) or `MiniMax-Hailuo-2.3-Fast` \|
116	\| `--first-frame <path-or-url>` \| string \| First frame image \|
117	\| `--callback-url <url>` \| string \| Webhook URL for completion \|
118	\| `--download <path>` \| string \| Save video to specific file \|
119	\| `--async` \| boolean \| Return task ID immediately \|
120	\| `--no-wait` \| boolean \| Same as `--async` \|
121	\| `--poll-interval <seconds>` \| number \| Polling interval (default: 5) \|
122
123	```bash
124	# Non-blocking: get task ID
125	mmx video generate --prompt "A robot." --async --quiet
126	# stdout: {"taskId":"..."}
127
128	# Blocking: wait and get file path
129	mmx video generate --prompt "Ocean waves." --download ocean.mp4 --quiet
130	# stdout: ocean.mp4
131	```
132
133	### video task get
134
135	Query status of a video generation task.
136
137	```bash
138	mmx video task get --task-id <id> [--output json]
139	```
140
141	### video download
142
143	Download a completed video by task ID.
144
145	```bash
146	mmx video download --file-id <id> [--out <path>]
147	```
148
149	---
150
151	### speech synthesize
152
153	Text-to-speech. Default model: `speech-2.8-hd`. Max 10k chars.
154
155	```bash
156	mmx speech synthesize --text <text> [flags]
157	```
158
159	\| Flag \| Type \| Description \|
160	\|---\|---\|---\|
161	\| `--text <text>` \| string \| Text to synthesize \|
162	\| `--text-file <path>` \| string \| Read text from file. Use `-` for stdin \|
163	\| `--model <model>` \| string \| `speech-2.8-hd` (default), `speech-2.6`, `speech-02` \|
164	\| `--voice <id>` \| string \| Voice ID (default: `English_expressive_narrator`)

People who liked this also viewed

6 picks

Type

minimax-multimodal-toolkit

39stars12forks

View on GitHub

How to install minimax-multimodal-toolkit?

nguyenphp/antigravity-marketing/minimax-multimodal-toolkit

$npx -y skills add nguyenphp/antigravity-marketing --skill minimax-multimodal-toolkit

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# MiniMax CLI — Agent Skill Guide
2
3	Use `mmx` to generate text, images, video, speech, music, and perform web search via the MiniMax AI platform.
4
5	## Prerequisites
6
7	```bash
8	# Install
9	npm install -g mmx-cli
10
11	# Auth (persisted to ~/.mmx/credentials.json)
12	mmx auth login --api-key sk-xxxxx
13
14	# Or pass per-call
15	mmx text chat --api-key sk-xxxxx --message "Hello"
16	```
17
18	Region is auto-detected. Override with `--region global` or `--region cn`.
19
20	---
21
22	## Agent Flags
23
24	Always use these flags in non-interactive (agent/CI) contexts:
25
26	\| Flag \| Purpose \|
27	\|---\|---\|
28	\| `--non-interactive` \| Fail fast on missing args instead of prompting \|
29	\| `--quiet` \| Suppress spinners/progress; stdout is pure data \|
30	\| `--output json` \| Machine-readable JSON output \|
31	\| `--async` \| Return task ID immediately (video generation) \|
32	\| `--dry-run` \| Preview the API request without executing \|
33	\| `--yes` \| Skip confirmation prompts \|
34
35	---
36
37	## Commands
38
39	### text chat
40
41	Chat completion. Default model: `MiniMax-M2.7`.
42
43	```bash
44	mmx text chat --message <text> [flags]
45	```
46
47	\| Flag \| Type \| Description \|
48	\|---\|---\|---\|
49	\| `--message <text>` \| string, required, repeatable \| Message text. Prefix with `role:` to set role (e.g. `"system:You are helpful"`, `"user:Hello"`) \|
50	\| `--messages-file <path>` \| string \| JSON file with messages array. Use `-` for stdin \|
51	\| `--system <text>` \| string \| System prompt \|
52	\| `--model <model>` \| string \| Model ID (default: `MiniMax-M2.7`) \|
53	\| `--max-tokens <n>` \| number \| Max tokens (default: 4096) \|
54	\| `--temperature <n>` \| number \| Sampling temperature (0.0, 1.0] \|
55	\| `--top-p <n>` \| number \| Nucleus sampling threshold \|
56	\| `--stream` \| boolean \| Stream tokens (default: on in TTY) \|
57	\| `--tool <json-or-path>` \| string, repeatable \| Tool definition JSON or file path \|
58
59	```bash
60	# Single message
61	mmx text chat --message "user:What is MiniMax?" --output json --quiet
62
63	# Multi-turn
64	mmx text chat \
65	--system "You are a coding assistant." \
66	--message "user:Write fizzbuzz in Python" \
67	--output json
68
69	# From file
70	cat conversation.json \| mmx text chat --messages-file - --output json
71	```
72
73	stdout: response text (text mode) or full response object (json mode).
74
75	---
76
77	### image generate
78
79	Generate images. Model: `image-01`.
80
81	```bash
82	mmx image generate --prompt <text> [flags]
83	```
84
85	\| Flag \| Type \| Description \|
86	\|---\|---\|---\|
87	\| `--prompt <text>` \| string, required \| Image description \|
88	\| `--aspect-ratio <ratio>` \| string \| e.g. `16:9`, `1:1` \|
89	\| `--n <count>` \| number \| Number of images (default: 1) \|
90	\| `--subject-ref <params>` \| string \| Subject reference: `type=character,image=path-or-url` \|
91	\| `--out-dir <dir>` \| string \| Download images to directory \|
92	\| `--out-prefix <prefix>` \| string \| Filename prefix (default: `image`) \|
93
94	```bash
95	mmx image generate --prompt "A cat in a spacesuit" --output json --quiet
96	# stdout: image URLs (one per line in quiet mode)
97
98	mmx image generate --prompt "Logo" --n 3 --out-dir ./gen/ --quiet
99	# stdout: saved file paths (one per line)
100	```
101
102	---
103
104	### video generate
105
106	Generate video. Default model: `MiniMax-Hailuo-2.3`. This is an async task — by default it polls until completion.
107
108	```bash
109	mmx video generate --prompt <text> [flags]
110	```
111
112	\| Flag \| Type \| Description \|
113	\|---\|---\|---\|
114	\| `--prompt <text>` \| string, required \| Video description \|
115	\| `--model <model>` \| string \| `MiniMax-Hailuo-2.3` (default) or `MiniMax-Hailuo-2.3-Fast` \|
116	\| `--first-frame <path-or-url>` \| string \| First frame image \|
117	\| `--callback-url <url>` \| string \| Webhook URL for completion \|
118	\| `--download <path>` \| string \| Save video to specific file \|
119	\| `--async` \| boolean \| Return task ID immediately \|
120	\| `--no-wait` \| boolean \| Same as `--async` \|
121	\| `--poll-interval <seconds>` \| number \| Polling interval (default: 5) \|
122
123	```bash
124	# Non-blocking: get task ID
125	mmx video generate --prompt "A robot." --async --quiet
126	# stdout: {"taskId":"..."}
127
128	# Blocking: wait and get file path
129	mmx video generate --prompt "Ocean waves." --download ocean.mp4 --quiet
130	# stdout: ocean.mp4
131	```
132
133	### video task get
134
135	Query status of a video generation task.
136
137	```bash
138	mmx video task get --task-id <id> [--output json]
139	```
140
141	### video download
142
143	Download a completed video by task ID.
144
145	```bash
146	mmx video download --file-id <id> [--out <path>]
147	```
148
149	---
150
151	### speech synthesize
152
153	Text-to-speech. Default model: `speech-2.8-hd`. Max 10k chars.
154
155	```bash
156	mmx speech synthesize --text <text> [flags]
157	```
158
159	\| Flag \| Type \| Description \|
160	\|---\|---\|---\|
161	\| `--text <text>` \| string \| Text to synthesize \|
162	\| `--text-file <path>` \| string \| Read text from file. Use `-` for stdin \|
163	\| `--model <model>` \| string \| `speech-2.8-hd` (default), `speech-2.6`, `speech-02` \|
164	\| `--voice <id>` \| string \| Voice ID (default: `English_expressive_narrator`)

People who liked this also viewed

6 picks

Type