Agent Guard

This is a multi-file skill: individual entry points intentionally implement only their documented role while the package provides the workflow as a whole. Scan AI agent skills, plugins, and MCP servers for malicious code BEFORE installation — catches prompt injection, credential theft, data exfiltration, and backdoors. Skills and the static MCP source scan use NVIDIA SkillSpector (static patterns + taint tracking + YARA + live OSV.dev CVE lookup + LLM semantic analysis, which runs by default through the user's own claude / opencode CLI login — no API key — or any hosted provider with a key; co

5 starsAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add elliottwaves-20/agent-guard --skill agent-guard

Manual install

git clone --depth 1 https://github.com/elliottwaves-20/agent-guard.git /tmp/agent-guard
mkdir -p ~/.claude/skills
cp -r /tmp/agent-guard/. ~/.claude/skills/agent-guard

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# agent-guard — scan first, install after

Scanner setup is an explicit bootstrap trust decision by the operator.
The setup scripts install trusted scanner tools and dependencies before any
target can be scanned; they do not constitute a prior security verdict for
the toolchain. Review setup separately. The scan-first workflow below applies
to target skills and packages, not to automatically provisioning the scanner.
Git installation requires the full scanned commit SHA. Package installation
examples require substituting the exact scanned package version; bare example
names are not installation approval. Existing destinations must be scanned
directly before reuse; the installer never silently substitutes their contents.

Skills and MCP servers are third-party code that runs with your user account's
permissions. A malicious one can read SSH keys, `.env` files, and browser
profiles, exfiltrate them, or hijack the agent itself through a poisoned
SKILL.md. This skill makes a security scan the mandatory first step of every
installation.

Because skills follow the open [SKILL.md standard](https://agentskills.io) and
MCP is an open protocol, the scan is agent-agnostic: one verdict covers Claude
Code, Codex, Antigravity, Gemini CLI, Hermes, OpenClaw, and any other compatible
agent — and the installer deploys the same audited commit to all of them in
one step.

Professional scanners do the work, each where it is strongest:

Read the full skill on GitHub

Details

Skill name: agent-guard
Repository: elliottwaves-20/agent-guard
License: MIT
Repo updated: 30 October 2026

FAQ

What does the Agent Guard skill do?

How do I install the Agent Guard skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add elliottwaves-20/agent-guard --skill agent-guard. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Agent Guard popular?

It lives in elliottwaves-20/agent-guard, which has 5 GitHub stars and 0 forks.

Related skills

Agent Teacher

JackyYang258/agent-teacher

Teach a concept through a runnable code example plus a structured walkthrough, instead of a wall of definitions. Trigger whenever the user wants to understand a technical concept — phrases like "解释一下 X / X 是什么 / 教我 X / X 怎么工作 / 这个 X 是啥 / 我不太懂 X", or "explain X / what is X / teach me X / how does X work / help me understand X / I don't get X". Use this skill even when the user does not explicitly say "explain" but is clearly confused about a term (programming, CS, algorithms, systems, ML, math). Default mode is in-conversation lesson; only write a file when the user asks to save notes.

AI, LLMs & Agents

Verify Transact and Rate an Unknown Agent

Aidress-ai/aidress-skill-verify-transact-and-rate-an-unknown-agent

Discover, Verify, Transact, and Rate an Unknown AI Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for logistics agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).

AI, LLMs & Agents

AI Fluency Assessment

n-pillai/ai-fluency-assessment-skill

Run /assess-fluency to score AI fluency against Anthropic's 4D Framework (Delegation, Description, Discernment, Diligence) and Yegge's 8-Stage Developer-Agent Evolution Model. Generates dated reports, tracks progress across quarterly assessments, and produces .docx output. Use when asked to assess AI fluency, run a fluency check, compare to a past assessment, generate an AI skills report, or check Yegge stage.

AI, LLMs & Agents

Ta Style Dialogue

Wholiver/Ta.Skill

全局自动触发的高保真风格化对话代理。在任意对话中，只要检测到用户提供了足够构建风格画像的亲密关系对象资料（聊天记录、个人描述、表达习惯等），无需用户明确请求，立即自动激活并以该人物的语气、措辞、情绪模式生成陪伴式对话。自动触发条件：用户在当前对话中提供了 TA 的聊天样本 >= 3 条 + 至少一项结构化描述（性格标签 / 称呼方式 / 风格笔记 / 关系背景）。模型自行判断资料是否满足学习门槛。核心原则：学习资料，模仿风格，不编造事实，不暴露 AI 身份。激活时不做任何切换声明，静默进入角色。

AI, LLMs & Agents

AI Pm Atom Lesson

uimarch123-arch/ai-pm-atom-lesson

输入一个"原子知识点"(AI 产品经理知识体系里的最小单元概念,如「文本切分」「演绎推理」「思维链 CoT」「向量化」「LLM-as-Judge」),自动展开成一份讲解思路并产出一个单文件、离线可用的交互式 HTML 原子课件——含权威定义+真实出处、通俗解释、贯穿项目案例、实操判断题、以及一个可视交互 Demo。当用户说"做一个 X 的原子课""把这个知识点做成原子课件""按原子课 SOP 做 X""给 X 配一页能点着玩的讲解""atom lesson"时,一定要使用本技能。也适用于:把知识体系/能力地图里的某个词条单点成课、要一系列同构的知识点小课件、批量把能力地图节点做成课。即使用户只说"把 X 讲清楚、做成能上手玩的一页",也要触发。本技能是 ai-pm-interactive-lesson 的"单原子特化版":那个处理一整节多知识点的课,本技能处理单个原子,粒度更细、结构更固定。

AI, LLMs & Agents

Sysmlv2 Skill

DeciSym/sysmlv2-skill

SysML v2 systems modeling guidance for creating valid models with correct syntax, proper type hierarchies, and validation workflows. Use when working with .sysml files, systems modeling, architecture documentation, SysON, or Eclipse SysML tools. Provides syntax reference, common patterns, error troubleshooting, and best practices from OMG specifications.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.