mcpguard-lite
Review MCP Configurations for Security Risks
MCP and agent configurations can contain risky permissions, commands, endpoints, or exposed credentials. MCPGuard Lite guides a redacted configuration review with evidence, heuristic warnings, and safer verification steps.
التثبيت باستخدام Agent لدي
انسخ هذا الطلب إلى Agent لديك. يتضمن صفحة Skill المعتمدة وملف manifest.
Review the Skillstore skill "mcpguard-lite" from https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite.md and its manifest at https://skillstore.io/api/skills/falconprojectai-arch-mcpguard-lite/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.يجب أن يواصل Agent عرض خطته وطلب أي تأكيد تفرضه سياسة الأمان.
موارد مهيّأة لـ Agents
استخدم هذه الروابط عندما يحتاج AI Agent أو crawler أو script إلى سياق نظيف بدلًا من قراءة الصفحة كاملة.
اختبرها
جارٍ استخدام "mcpguard-lite". A redacted MCP configuration declares a shell command that downloads a remote script and immediately runs it.
النتيجة المتوقعة:
- Observation: The declared command combines a remote download with immediate shell execution.
- Heuristic warning: The downloaded code could change, and the configuration does not establish its behavior.
- Verification: Review the referenced source and determine whether a pinned, verified release can replace the download-and-run pattern.
- Review boundary: No command was executed and no code was downloaded.
جارٍ استخدام "mcpguard-lite". A redacted agent configuration grants wildcard file permissions and references an environment variable named API_TOKEN.
النتيجة المتوقعة:
- Observation: The configuration declares wildcard file permissions and references API_TOKEN.
- Heuristic warning: Wildcard permissions may exceed the intended scope; the variable name alone does not prove credential exposure.
- Verification: Confirm required file access and narrow permissions where possible. Verify the variable source without sharing its value.
- False-positive possibility: The environment reference may be intentional and appropriate.
التدقيق الأمني
آمنThe static catalog contains no findings, so no per-finding verdicts are required. No evidence found of prompt injection, credential collection, or unauthorized execution in the supplied skill. Its instructions require redacted input, prohibit executing configurations, and distinguish heuristic warnings from confirmed observations.
شارك واستشهد بهذا التقرير
شارك تقرير التقييم المرتبط بالإصدار والشارة المحايدة وبطاقة التضمين والاستشهادات. تعرض Skillstore الأدلة من دون أن تقرر ما إذا كانت هذه المهارة آمنة.
نسخ رابط التقرير
https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportشارة Markdown
[](https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite?utm_source=security_passport_badge)شارة HTML
<a href="https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/falconprojectai-arch-mcpguard-lite/security.svg" alt="Skillstore security assessment" loading="lazy"></a>بطاقة قابلة للتضمين
<iframe src="https://skillstore.io/embed/skills/falconprojectai-arch-mcpguard-lite.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>الاستشهادات الأكاديمية (APA · BibTeX · CFF)
اقتباس APA
falconprojectai-arch. (2026). mcpguard-lite security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite/audits/1اقتباس BibTeX
@techreport{falconprojectai-arch-falconprojectai-arch-mcpguard-lite-2026,
author = {falconprojectai-arch},
title = {mcpguard-lite security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "mcpguard-lite security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "falconprojectai-arch"
date-released: "2026-11-12"
url: "https://skillstore.io/skills/falconprojectai-arch-mcpguard-lite/audits/1"
identifiers:
- type: other
value: "skillstore:falconprojectai-arch-mcpguard-lite:audit:1"
description: "Skillstore immutable audit report identifier"
تقييم Skillstore
سبب هذا التقييم موثوقية الأدلة: متوسطما الذي يمكنك بناؤه
Review a Developer Configuration
Inspect a redacted MCP configuration for commands, remote endpoints, and permission patterns before enabling its servers.
Triage Configuration Concerns
Distinguish observable configuration risks from heuristic warnings and identify questions requiring manual verification.
Check Team Configuration Templates
Review shared agent settings for wildcard permissions, insecure transport hints, and credential-handling concerns before distributing them.
جرّب هذه الموجّهات
Review this redacted MCP JSON configuration: [paste configuration]. List potential security concerns and safer verification steps. Do not execute commands or reproduce secrets.
Inspect this redacted agent configuration: [paste configuration]. Identify commands, arguments, environment variable names, endpoints, and transport settings. Explain suspicious patterns without executing anything.
Review this redacted configuration: [paste configuration]. Focus on permission scopes, wildcard access, credential-like values, and unexplained environment references. Quote only non-secret fragments. Suggest verification steps.
Review this redacted configuration: [paste configuration]. Separate confirmed observations, heuristic warnings, and possible false positives. Explain each concern and propose safer verification steps. State what configuration text cannot establish. Do not execute commands or download code.
أفضل الممارسات
- Remove secrets before sharing configuration text, while preserving relevant structure and environment variable names.
- Verify flagged commands, endpoints, and permissions through trusted documentation without executing the configuration during review.
- Treat observations and heuristic warnings separately, and rotate real credentials when exposure has occurred.
تجنب
- Submitting live API keys, tokens, passwords, or production credentials for inspection.
- Executing configuration commands or downloading referenced code to investigate a warning.
- Treating a clean review as proof of safety or a heuristic warning as proof of malicious behavior.
الأسئلة المتكررة
What input should I provide?
Does the skill run my configuration?
Does a clean review guarantee safety?
Does every warning indicate malicious behavior?
How should exposed credentials be handled?
Can I use this with Claude, Codex, or Claude Code?
تفاصيل المطور
المؤلف
falconprojectai-archالترخيص
غير معروف
مراجعة Skillstore
r1
تنبيه الإصدار
لم يعلن المؤلف عن إصدار.
مرجع
46ef99ad4e3887c3e9047faa4acfaf85100cc124
حداثة الصيانة
١٢/١١/٢٠٢٦
الاستخدام
0 تنزيلات · 0 مشاهدات
بنية الملفات
📄 SKILL.md