此内容尚未提供您的语言版本,正在以英文显示。

Gws Modelarmor Sanitize Prompt

技能已验证活跃

Google Model Armor: Sanitize a user prompt through a Model Armor template.

目的

To enhance AI safety by cleaning user prompts against predefined Model Armor templates before they are processed, preventing potential misuse or harmful content.

功能

Sanitize user prompts via Model Armor templates
Accepts text input or full JSON request body
Supports reading input from stdin
Integrates with Google Cloud Model Armor

使用场景

When processing potentially untrusted user input that needs to be validated against a security template.
As a pre-processing step in AI agent workflows to ensure prompt safety.
To adhere to Google Cloud Model Armor policies for prompt handling.

非目标

Sanitizing model responses (use `+sanitize-response` instead)
Providing AI Model Armor template management beyond basic usage
Operating independently of the `gws` CLI

安装

npx skills add googleworkspace/cli

通过 npx 运行 Vercel skills CLI(skills.sh)— 需要本地安装 Node.js,以及至少一个兼容 skills 的智能体(Claude Code、Cursor、Codex 等)。前提是仓库遵循 agentskills.io 格式。

质量评分

已验证

99 /100

1 day ago 分析

信任信号

最近提交2 days ago

GitHub 所有者 googleworkspace

星标26.2k

许可证Apache-2.0

网站developers.google.com

状态

查看源代码

类似扩展

Fixflow

使用严格的交付工作流执行编码任务：构建完整计划、分步实现、持续运行测试，并默认在每一步 (`per_step`) 后提交。当用户要求行为驱动交付或需求不明确时，支持显式提交策略覆盖 (`final_only`, `milestone`) 和可选的 BDD（给定/当/则）。

Trader Regime

Detect current market regime using npx neural-trader — bull/bear/ranging/volatile classification with recommended strategy

Setup

Use first for install/update routing — sends setup, doctor, or MCP requests to the correct OMC setup flow

Project Session Manager

Worktree-first dev environment manager for issues, PRs, and features with optional tmux sessions

Mcp Setup

Configure popular MCP servers for enhanced agent capabilities

Oh My Claudecode

Process-first advisor routing for Claude, Codex, or Gemini via `omc ask`, with artifact capture and no raw CLI assembly