vwu.ai MiniMax Models

Call minimax models MiniMax-Hailuo series on vwu.ai with OpenAI-compatible chat completions using your configured API key.

Security Scan

The skill appears to implement what it claims (calling vwu.ai MiniMax models) but has inconsistencies and small privacy/security issues: the registry metadata doesn't declare the required API key or required CLI tools, and the script unnecessarily echoes the first 8 characters of the API key on error.

Purpose & Capability

The skill's name, SKILL.md, models.txt, and vwu-chat.sh all align: they call vwu.ai's OpenAI-compatible chat completions for MiniMax models. However the registry metadata lists no required env vars or primary credential even though both the README and the script require VWU_API_KEY (and optionally VWU_BASE_URL). This metadata omission is an inconsistency that should be corrected.

Instruction Scope

SKILL.md instructs only to set VWU_API_KEY and how to call vwu-chat; the included vwu-chat.sh uses only VWU_API_KEY and an optional VWU_BASE_URL and calls the vwu.ai /v1/chat/completions endpoint. It does not read unrelated system paths. Concern: on API error the script prints a partially redacted API key (${VWU_API_KEY:0:8}***), which leaks the first 8 characters of the key to console logs/errors — unnecessary and increases risk if logs are collected.

Install Mechanism

There is no install spec (instruction-only), which is low risk. However the provided script assumes certain CLI tools are available (zsh shebang, curl, jq, grep, sed). The registry metadata lists no required binaries; that mismatch could cause runtime failures and should be declared.

Credentials

Requesting an API key for vwu.ai is proportional and expected. But the skill metadata does not declare VWU_API_KEY as required/primary credential, creating an unclear trust surface. Additionally, the script's partial-key echo on error leaks a secret fragment to stdout/stderr — not justified by the task.

Persistence & Privilege

The skill does not request persistent presence, does not set always:true, and does not modify other skill/system configs. It runs a simple CLI wrapper and exits; no privilege escalation or persistent agents are present.

What to consider before installing

This skill generally does what it claims (wraps vwu.ai chat completions), but there are some issues to consider before installing: 1) The registry metadata omits the required VWU_API_KEY and doesn't list required CLI tools (zsh, curl, jq, grep, sed). Expect to set VWU_API_KEY in your environment and ensure the tools are installed. 2) The script prints the first 8 characters of your API key on API errors — avoid using a sensitive/reused key or ensure logs are not exposed; ask the author to remove that behavior. 3) Confirm you trust vwu.ai and the skill source (there is no homepage/source link). If you need stronger assurance, request the publisher add explicit required-env and required-binaries metadata, remove the API-key echo, and publish a trustworthy source/homepage before use.

minimax Models for vwu.ai

vwu.ai 平台上的 minimax 模型调用技能。

支持的模型

• MiniMax-Hailuo-02
• MiniMax-Hailuo-2.3
• MiniMax-Hailuo-2.3-Fast

共 3 个模型。

配置

使用前需要设置 vwu.ai API key:

export VWU_API_KEY="your-key-here"

获取 key: https://vwu.ai 控制台

使用示例

# 调用模型
vwu-chat --model MiniMax-Hailuo-02 "你的问题"

API 兼容性

所有模型兼容 OpenAI API 格式，支持标准 chat completions 接口。

额度提示

如果使用时提示"额度不足"或类似错误：

1. 访问 https://vwu.ai
2. 在控制台充值或调整额度
3. 或生成新的 API key

---

由 Claw 自动生成 | 源数据: vwu.ai