aguara MCP工具 是 AI Skill Hub 本期精选MCP工具之一。综合评分 8.0 分,整体质量较高。我们强烈推荐将其纳入你的 AI 工具库,帮助提升工作效率。
专为AI代理和MCP服务器设计的开源安全扫描工具。通过静态分析检测技能配置漏洞、数据泄露风险和安全隐患,帮助开发者和安全团队加强AI系统防护。适合关注AI安全的开发者、DevSecOps工程师和企业安全团队使用。
aguara MCP工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
专为AI代理和MCP服务器设计的开源安全扫描工具。通过静态分析检测技能配置漏洞、数据泄露风险和安全隐患,帮助开发者和安全团队加强AI系统防护。适合关注AI安全的开发者、DevSecOps工程师和企业安全团队使用。
aguara MCP工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/garagon/aguara
# 方式二:手动配置 claude_desktop_config.json
{
"mcpServers": {
"aguara-mcp--": {
"command": "npx",
"args": ["-y", "aguara"]
}
}
}
# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
# 安装后在 Claude 对话中直接使用 # 示例: 用户: 请帮我用 aguara MCP工具 执行以下任务... Claude: [自动调用 aguara MCP工具 MCP 工具处理请求] # 查看可用工具列表 # 在 Claude 中输入:"列出所有可用的 MCP 工具"
// claude_desktop_config.json 配置示例
{
"mcpServers": {
"aguara_mcp__": {
"command": "npx",
"args": ["-y", "aguara"],
"env": {
// "API_KEY": "your-api-key-here"
}
}
}
}
// 保存后重启 Claude Desktop 生效
Check what your AI agents are about to trust.
Open-source security engine for agent instructions, tool configurations, and software dependencies.
<p align="center"> <a href="https://github.com/garagon/aguara/actions/workflows/ci.yml"><img src="https://github.com/garagon/aguara/actions/workflows/ci.yml/badge.svg" alt="CI"></a> <a href="https://codecov.io/gh/garagon/aguara"><img src="https://codecov.io/gh/garagon/aguara/branch/main/graph/badge.svg" alt="Coverage"></a> <a href="https://goreportcard.com/report/github.com/garagon/aguara"><img src="https://goreportcard.com/badge/github.com/garagon/aguara" alt="Go Report Card"></a> <a href="https://pkg.go.dev/github.com/garagon/aguara"><img src="https://pkg.go.dev/badge/github.com/garagon/aguara.svg" alt="Go Reference"></a> <a href="https://github.com/garagon/aguara/releases"><img src="https://img.shields.io/github/v/release/garagon/aguara" alt="GitHub Release"></a> <a href="LICENSE"><img src="https://img.shields.io/github/license/garagon/aguara" alt="License"></a> </p>
A project brings more than source code into your environment. Its dependencies can run install scripts. Its agent settings can approve commands or start tools. Its instructions can ask an agent to read credentials or execute a helper you have not reviewed.
Aguara checks those files for known-malicious packages, suspicious behavior, and risky permissions before you use them. The report points you to the affected package or file so you can investigate before installing dependencies, running CI, or giving an agent access to the project.
Use it when evaluating a repository or skill, reviewing a change, or adding a security check to an agent workflow. You do not need an AI agent to use its package and code checks.
Analysis runs locally. No package execution, content upload, telemetry, or LLM calls.
Quick start | Coverage | CI | Integrations | Development | Security
go install github.com/garagon/aguara/cmd/aguara@v0.28.0 ```
Release binaries are available for Linux, macOS, and Windows on the Releases page. The container supports Linux amd64 and arm64. Go installs do not inject release version metadata; use release artifacts when that metadata or signature verification is required.
The following alternatives install the published release:
```bash
docker run --rm -v "$PWD:/repo:ro" ghcr.io/garagon/aguara:0.28.0 audit /repo
Install the published release:
curl -fsSL https://raw.githubusercontent.com/garagon/aguara/v0.28.0/install.sh \
| VERSION=v0.28.0 sh
The default location is ~/.local/bin; add it to your PATH if needed. The installer verifies the archive checksum. See signature verification for verifying its signing identity, or other installation options.
From the project directory, run:
aguara audit .
This combines a malicious-package check with a content scan. It does not install dependencies or run the project's commands. Read the findings before taking the next step; a passing result is not a guarantee that the project is safe.
Trust boundary: in v0.28.0, audit, CI scans and the public scanning API ignore target-owned exclusions and suppressions by default. Local scan can still honor them; use --project-policy ignore for unfamiliar projects. Older releases may trust repository-owned policy. See the upgrade notes before changing an existing integration.
Lockfiles let Aguara check resolved dependencies without installing them. Coverage depends on what the file records and what the active intelligence snapshot knows.
| Ecosystem | Evidence read |
|---|---|
| npm | node_modules, pnpm store, package-lock.json, pnpm-lock.yaml, classic and Berry yarn.lock, text bun.lock |
| PyPI | Installed site-packages, .pth files, supported cache locations |
| Go | go.sum, go.mod |
| Rust | Public-registry entries in Cargo.lock |
| PHP | composer.lock |
| Ruby | Gemfile.lock |
| Java | pom.xml, Gradle lockfiles |
| .NET | packages.lock.json, supported project files |
Matching supports exact versions and advisories affecting every version of a package. Bounded version ranges are supported for npm semver, not arbitrary ranges in every ecosystem. This is malicious-package detection, not comprehensive CVE coverage.
Unambiguous npm: aliases resolve to the real package in package-lock, pnpm, Yarn classic, Yarn Berry, and Bun. Classic Yarn uses the alias target name and the block's resolved version; conflicting descriptors and ambiguous non-registry identities are skipped rather than assigned a guessed package identity. Binary bun.lockb is not parsed; a repository with only that file returns an error asking for text bun.lock.
Lockfile and dependency-manifest inputs must be regular files no larger than 50 MiB. Leaf symlinks, including dangling links, return an error rather than being followed or silently skipped. This read boundary does not confine parent directories or guarantee that file contents remain unchanged during a check. Installed-package scanning is a separate path.
Legacy package-lock.json dependency trees have a 128-level nesting limit and a 50 MiB cumulative subtree-decoding budget, in addition to the file-size limit. Exceeding either returns an error, not a partial successful package check.
scan offers terminal, JSON, SARIF, and Markdown output. audit offers a combined terminal or JSON report. Use SARIF from scan for GitHub Code Scanning; other SAST services may require a different schema.
The Go library exposes the same content-scanning engine for tools that need analysis without a CLI subprocess:
package main
import (
"context"
"fmt"
"log"
"github.com/garagon/aguara"
)
func main() {
result, err := aguara.ScanContent(context.Background(), "Content to inspect", "skill.md")
if err != nil {
log.Fatal(err)
}
for _, finding := range result.Findings {
fmt.Printf("%s %s:%d\n", finding.RuleID, finding.FilePath, finding.Line)
}
}
Scan accepts a file or directory; ScanContent accepts content directly without loading a file. ListRules and ExplainRule expose the catalog. Pin the Go module version in your application. Consumers can call the engine during a live workflow, but interception, authorization, session policy, and enforcement remain the consumer's responsibility.
Aguara MCP is a separate integration that exposes scanning tools to agents. Aguara itself does not require an MCP server, hosted account, or model provider.
針對AI系統安全的专业工具,填补MCP服务器安全检测空白。代码质量好,社区活跃度有待提升。Go语言性能优秀,值得AI安全践行者采用。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
经综合评估,aguara MCP工具 在MCP工具赛道中表现稳健,质量优秀。如果你已有明确的使用需求,可以直接上手体验;如果还在评估阶段,建议对比同类工具后再做决策。
| 原始名称 | aguara |
| 原始描述 | 开源MCP工具:Security scanner for AI agent skills and MCP servers. Static analysis, incident 。⭐77 · Go |
| Topics | 安全扫描MCP服务器AI代理静态分析数据泄露检测DevSecOps |
| GitHub | https://github.com/garagon/aguara |
| License | Apache-2.0 |
| 语言 | Go |
收录时间:2026-05-18 · 更新时间:2026-05-19 · License:Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端