Token导航 LogoToken导航TokenDH.com
研究检索可写文件clawhub未标认证来源可访问clear审计通过

memguardmemguard 搜索

Agent Skill

memguard 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

2,252

周安装

92

GitHub Stars

公开资料未说明

下载量

729
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:memguard(memguard 搜索)
来源仓库:https://github.com/suhteevah/memguard
安装命令:
openclaw skills install memguard
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install memguard

简介

memguard 检测内存泄漏与资源管理问题,如未关闭句柄、循环引用等。

  • 适用于 Node.js 或 JavaScript 项目的运行时健康检查。
  • 输出报告包含潜在风险点与修复建议,供开发者参考。
  • 不覆盖所有边缘 case,需结合日志与调试工具综合排查。
  • 建议在开发阶段定期运行以预防生产环境问题。memguard 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

name
memguard
description
Memory leak & resource management scanner -- detects unclosed handles, event listener leaks, circular references, unbounded caches, missing cleanup, dangling timers, and resource lifecycle issues across all languages
homepage
https://memguard.pages.dev
metadata
user-invocable
true
disable-model-invocation
false

MemGuard -- Memory Leak & Resource Management Scanner

MemGuard scans codebases for memory leaks and resource management issues: unclosed file handles, event listener leaks, circular references, unbounded caches/maps, missing cleanup in React useEffect, dangling timers/intervals, unreleased database connections, missing stream close, forgotten subscriptions, and retained DOM references -- across JS/TS, Python, Java, Go, Rust, and C#. It uses regex-based pattern matching against 90+ resource lifecycle patterns, lefthook for git hook integration, and produces markdown reports with actionable remediation recommendations. 100% local. Zero telemetry.

Commands

Free Tier (No license required)

memguard scan [file|directory]

One-shot memory leak and resource management scan of files or directories.

How to execute:

bash "<SKILL_DIR>/scripts/memguard.sh" scan [target]

What it does:

  1. Accepts a file path or directory (defaults to current directory)
  2. Discovers all source files (skips .git, node_modules, binaries, images, .min.js)
  3. Runs 90+ resource lifecycle detection patterns against each file
  4. Respects .gitignore and allowlist files
  5. Calculates a resource health score (0-100) per file and overall
  6. Grades: A (90-100), B (80-89), C (70-79), D (60-69), F (<60)
  7. Outputs findings with: file, line number, check ID, severity, description, recommendation
  8. Exit code 0 if score >= 70, exit code 1 if too many issues found
  9. Free tier limited to 5 files per scan

Example usage scenarios:

  • "Scan my code for memory leaks" -> runs memguard scan .
  • "Check this file for resource leaks" -> runs memguard scan src/app.ts
  • "Find unclosed file handles in my project" -> runs memguard scan src/
  • "Are there event listener leaks in my codebase?" -> runs memguard scan .
  • "Check for missing useEffect cleanup" -> runs memguard scan .

memguard hook install

Install git pre-commit hooks that scan staged files for memory leaks before every commit.

How to execute:

bash "<SKILL_DIR>/scripts/memguard.sh" hook install

What it does:

  1. Copies lefthook config to project root
  2. Installs lefthook pre-commit hook
  3. On every commit: scans all staged files for resource leaks, blocks commit if critical/high findings, shows remediation advice

memguard hook uninstall

Remove MemGuard git hooks.

bash "<SKILL_DIR>/scripts/memguard.sh" hook uninstall

memguard report [directory]

Generate a markdown resource health report with findings, severity breakdown, and remediation steps.

bash "<SKILL_DIR>/scripts/memguard.sh" report [directory]

What it does:

  1. Runs full scan of the directory
  2. Generates a formatted markdown report from template
  3. Includes per-file breakdowns, resource health scores, remediation priority
  4. Output suitable for code reviews and tech debt tracking

Pro Tier ($19/user/month -- requires MEMGUARD_LICENSE_KEY)

memguard watch [directory]

Continuous monitoring mode that watches for file changes and re-scans automatically.

How to execute:

bash "<SKILL_DIR>/scripts/memguard.sh" watch [directory]

What it does:

  1. Validates Pro+ license
  2. Monitors directory for file changes using filesystem polling
  3. Re-scans changed files automatically
  4. Outputs real-time findings as files are modified
  5. Useful during development for immediate feedback on resource leaks

memguard ci [directory]

CI/CD integration mode with structured exit codes and machine-readable output.

bash "<SKILL_DIR>/scripts/memguard.sh" ci [directory]

What it does:

  1. Validates Pro+ license
  2. Runs full scan of the directory
  3. Outputs findings in a CI-friendly format
  4. Exit code 0 if clean, 1 if critical/high issues found
  5. Compatible with GitHub Actions, GitLab CI, Jenkins, CircleCI
  6. Supports severity threshold configuration

Team Tier ($39/user/month -- requires MEMGUARD_LICENSE_KEY with team tier)

memguard team-report [directory]

Aggregate team metrics across the codebase with trend analysis.

bash "<SKILL_DIR>/scripts/memguard.sh" team-report [directory]

What it does:

  1. Validates Team+ license
  2. Runs comprehensive scan of the entire codebase
  3. Generates aggregate team metrics and category breakdowns
  4. Identifies hotspot files with the most resource issues
  5. Produces trend data when baselines exist for comparison
  6. Output suitable for engineering leadership and sprint planning

memguard baseline [directory]

Establish a baseline of known resource issues for tracking progress over time.

bash "<SKILL_DIR>/scripts/memguard.sh" baseline [directory]

What it does:

  1. Validates Team+ license
  2. Scans directory and records all current findings as baseline
  3. Saves baseline to .memguard-baseline.json
  4. Future scans report improvement/regression against the baseline
  5. Useful for tech debt tracking in established codebases

memguard status

Show license and configuration information.

bash "<SKILL_DIR>/scripts/memguard.sh" status

Detected Resource Leak Patterns

MemGuard detects 90+ resource lifecycle patterns across 6 categories:

CategoryExamplesSeverity
File Handles (FH)Python open() without close/with, Java FileInputStream without close, Node fs.open without fs.close, missing using/with statements, unclosed database connectionsCritical/High
Event Listeners (EL)addEventListener without removeEventListener, on() without off(), missing event cleanup in React lifecycle, event emitter leaks, global event handlers never removedHigh/Medium
Circular References (CR)Parent-child circular refs, self-referencing objects, closure-captured references to parent scope, mutual references between classes, WeakRef/WeakMap needed but not usedHigh/Medium
Unbounded Caches (UC)Map/Set growing without limit, in-memory cache without TTL/eviction, global arrays accumulating, memoization without size limit, localStorage unbounded writesHigh/Medium
React Cleanup (RC)useEffect without cleanup return, setInterval in useEffect without clearInterval, subscription in useEffect without unsubscribe, state update after unmount, ref held after unmountHigh/Medium
Timers & Streams (TM)setInterval without clearInterval, setTimeout in loops, stream pipe without error handling, readable stream not destroyed, missing stream.end(), unclosed WebSocketCritical/High

Configuration

Users can configure MemGuard in ~/.openclaw/openclaw.json:

{
  "skills": {
    "entries": {
      "memguard": {
        "enabled": true,
        "apiKey": "YOUR_LICENSE_KEY_HERE",
        "config": {
          "severityThreshold": "high",
          "ignorePatterns": ["**/test/**", "**/fixtures/**", "**/*.test.*"],
          "ignoreChecks": [],
          "allowlistFile": ".memguard-allowlist",
          "reportFormat": "markdown"
        }
      }
    }
  }
}

Important Notes

  • Free tier works immediately with no configuration
  • All scanning happens locally -- no code is sent to external servers
  • License validation is offline -- no phone-home or network calls
  • Pattern matching only -- no AST parsing, no external dependencies beyond bash
  • Supports scanning JS/TS, Python, Java, Go, Rust, and C# in a single pass
  • Git hooks use lefthook which must be installed (see install metadata above)
  • Exit codes: 0 = clean (score >= 70), 1 = resource issues detected (for CI/CD integration)
  • Scoring: critical=25pts, high=15pts, medium=8pts, low=3pts deducted from 100 base

Error Handling

  • If lefthook is not installed and user tries hook install, prompt to install it
  • If license key is invalid or expired, show clear message with link to https://memguard.pages.dev/renew
  • If a file is binary, skip it automatically with no warning
  • If no scannable files found in target, report clean scan with info message
  • If .memguard-allowlist is missing, skip allowlist filtering gracefully

When to Use MemGuard

The user might say things like:

  • "Scan my code for memory leaks"
  • "Find unclosed file handles in my project"
  • "Check for event listener leaks"
  • "Are there missing useEffect cleanups?"
  • "Find resource leaks in my codebase"
  • "Scan for dangling timers and intervals"
  • "Check for unbounded caches"
  • "Find circular references in my code"
  • "Detect missing stream close calls"
  • "Check for database connection leaks"
  • "Run a memory scan before I push"
  • "Set up pre-commit hooks for resource leak detection"
  • "Generate a resource health report"
  • "Find forgotten event subscriptions"
  • "Check for missing cleanup functions"
  • "Scan for retained DOM references"
  • "Establish a baseline for resource issues"
  • "Monitor my code for memory leaks in real-time"

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

82.65%
按下载量换算603

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

可写文件

该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills