MessageGuard: Outgoing Message Filter Skill
**Purpose**: MessageGuard filters outgoing text to prevent secret leaks and sensitive data exposure by using pattern-based detection and configurable actions (mask, block, or warn).
Best use case
MessageGuard: Outgoing Message Filter Skill is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
**Purpose**: MessageGuard filters outgoing text to prevent secret leaks and sensitive data exposure by using pattern-based detection and configurable actions (mask, block, or warn).
Teams using MessageGuard: Outgoing Message Filter Skill should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/messageguard/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How MessageGuard: Outgoing Message Filter Skill Compares
| Feature / Agent | MessageGuard: Outgoing Message Filter Skill | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
**Purpose**: MessageGuard filters outgoing text to prevent secret leaks and sensitive data exposure by using pattern-based detection and configurable actions (mask, block, or warn).
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Marketing
Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.
AI Agents for Startups
Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
SKILL.md Source
### MessageGuard: Outgoing Message Filter Skill **Purpose**: MessageGuard filters outgoing text to prevent secret leaks and sensitive data exposure by using pattern-based detection and configurable actions (mask, block, or warn). ### Advanced Configuration Options: - **`mode`**: Determines the global action for matched patterns. Options are: - `mask`: Replace sensitive data with the `mask_char`. - `block`: Prevent the message from being sent entirely. - `warn`: Allow the message but generate warnings. - **`mask_char`**: The character(s) used to replace sensitive content when `mode` is set to `mask`. - **`patterns`**: Define or customize regex-based detections. Built-ins exist for API keys, credentials, and more (e.g., AWS keys, JWTs). Add new patterns based on your requirements. - **`logging`**: Enable detections to be logged as structured JSON for monitoring, debugging, or compliance needs. Configure the `log_path` for the output location. - **Custom Patterns**: Users can define their custom patterns to override built-ins or extend functionalities. This supports regex and granular action control (mask, block, warn). **Installation** 1. Clone the repository: `git clone git@github.com:AndrewAndrewsen/MessageGuard.git`. 2. Navigate to the directory. The skill is dependency-free, relying only on the Python standard library.
Related Skills
jiuwu-message
调用久吾消息网关HTTP接口给企业内部联系人发送消息。使用场景：(1) 需要向企业内部同事发送通知或提醒时，(2) 调用时传入接收人工号(code)、消息内容(text)和标题(title)
Feishu Message Skill
A unified toolkit for Feishu messaging operations, providing a single CLI entry point for common tasks.
message-chunker
智能消息分段工具。自动将长消息拆分成多条发送，避免飞书API消息长度限制。当需要发送长消息（如汇报、方案、日志）时自动激活。
send-message
Send and receive direct messages on OpenAnt. Use when the agent needs to communicate privately with another user, check for new messages, read conversations, reply to someone, or start a chat. Covers "message someone", "send a DM", "reply to", "read messages", "check conversations", "any new messages?", "what did they say?", "check inbox".
medication-adherence-message-gen
Use medication adherence message gen for academic writing workflows that need structured execution, explicit assumptions, and clear output boundaries.
lipinski-rule-filter
Filter compound libraries based on Lipinski's Rule of Five for drug-likeness.
stock-filter-skills
股票多条件筛选、热门因子管理、Jiuyan 数据查询和抖音热点分析。提供 17 个 CLI 工具覆盖四大模块。
message-friendly
Optimizes Markdown-formatted messages for IM platforms (WeChat, DingTalk, etc.) by removing unsupported markup and using clean text layouts.
message-split
长消息自动拆分 skill。当回复内容超过渠道限制时，自动将消息拆分为多条有序发送，避免截断或丢消息。Auto-split long messages into smaller chunks before sending, with sequence markers.
competitor-message-differ
比较竞品在信息架构和叙事上的差异，找出可借鉴点与应避免模仿点。；use for competitor, messaging, positioning workflows；do not use for 抄袭竞品文案, 编造竞品数据.
---
name: article-factory-wechat
humanizer
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.