Skip to content
AgentGuard

Your AI CISOSecure EnterpriseAt Every Layer

Empower Development And Security Teams With Agentic AI To Find And Fix Risks Across Code, Applications, AndInfrastructure, While Securing AI Systems And Agents From Development To Deployment.

＋ Open Source ＋＋ Powered By GoPlus ＋＋ Local First ＋
Start FreeBook a Demo

Deploy AgentGuard

curl -fsSL https://agentguard.gopluslabs.io/install.sh | bash

For macOS / Linux: installs the OSS CLI and enables local-first runtime protection where the supported connection path exposes it.

10,000+
Actions Evaluated
500+
Threats Blocked
6
Policy Rules
<50ms
Avg Decision Time
The Risk Moved Into Action

Why AI Agent Security Matters

AI Agents Do More Than Generate Text. They Call Tools, Access Data, Execute Commands, And Interact With External Services. Each Action Creates A Security Decision That Text-Only Controls May Not Cover.

  • Agents Can Take Real Actions: Commands, File Changes, API Calls And Automated Workflows Can Create Immediate Consequences.
  • Skills, Plugins And MCP Servers Add Supply-Chain Risk: Unreviewed Skills, Plugins And MCP Servers Can Introduce Prompt Injection, Malicious Tools, Credential Leaks Or Backdoors.
  • Prompt Injection Can Influence Tool Behavior: Manipulated Instructions Can Change Which Tools An Agent Chooses And How Those Tools Are Used.
  • Static Scanning Is Not Enough: Components Can Be Reviewed Before Use, While High-Impact Actions Still Need A Runtime Decision.

Agents Can Take Real Actions

Commands, File Changes, API Calls And Automated Workflows Can Create Immediate Consequences.

Skills, Plugins And MCP Servers Add Supply-Chain Risk

Unreviewed Skills, Plugins And MCP Servers Can Introduce Prompt Injection, Malicious Tools, Credential Leaks Or Backdoors.

Prompt Injection Can Influence Tool Behavior

Manipulated Instructions Can Change Which Tools An Agent Chooses And How Those Tools Are Used.

Static Scanning Is Not Enough

Components Can Be Reviewed Before Use, While High-Impact Actions Still Need A Runtime Decision.

AgentGuard Evaluates Documented Shell, File, Tool, Network, Secret, Sensitive-Write, And Webhook-Exfiltration Action Categories Before Execution.

Runtime Decisions & Workflow

Evaluate High-Risk
Agent Actions Before
They Execute

Runtime policy decision workflow
Protection Coverage

Three Layers Of Protection
For AI Agents — AI

Runtime protection, deep security analysis, and continuous environment monitoring.

Runtime Guard

Evaluate Documented High-Risk Action Categories Before Execution
  • Shell Commands
  • File Access
  • Tool Actions
  • Network Requests
  • Secret Access
  • Sensitive Writes
  • Webhook Exfiltration
Explore Runtime Guard

Deep Scan

Review Agent-Related Components For Documented Security Risks Before You Trust Them.
  • Skills
  • Plugins
  • Agents
  • MCP Servers
  • Repositories, Packages And URLs
  • Through Documented API Groups
Explore Deep Scan

AI Agent Environment Patrol

Monitor Documented Security-Relevant Changes Inside An AI Agent Workspace
  • Suspicious Skills
  • Modified Plugins
  • New MCP Servers
  • Drift In Trusted Files
Explore AI Agent Patrol

Six Security Detectors. One Scan.

Credential Exposure Detection

Identify Embedded Secrets, Tokens, Private Keys, And Connection Strings.

Prompt Injection Detection

Review Encoded, Obfuscated, Role-Manipulation, And System-Prompt Extraction Patterns.

Malicious Command Detection

Identify Command Patterns Associated With Remote Execution And Unsafe Payload Delivery.

Data Exfiltration Detection

Review Suspicious Access And Transfer Paths That May Expose Sensitive Data.

Permission Abuse Analysis

Compare Declared Tool Needs With Risky Permission Combinations.

URL Analysis

Review Suspicious Domains, Phishing Patterns, Shortened Links, And Lookalike URLs.

See AgentGuard in Real
Agent Workflows

Open The Quickstart
Documented Workflow Preview
01

A documented agent workflow requests an action or component scan. Input and integration depth must be verified in the target environment.

Protect High-Impact Coding-Agent Actions

Evaluate Documented Shell, File, Network, Secret, Write, And Tool-Action Categories Before Execution Through A Supported Integration Path.

For Developers

Review Skills, Plugins And MCP Components

Scan Documented Component Types Before Adding Them To A Trusted Agent Workflow.

Explore MCP Security

Give AI CISOs And Security Teams Verifiable Evidence

Review Documented Findings, Qualified Audit Events, Integration Boundaries, And Public Advisories When Evaluating AI CISO Responsibilities And Wider Deployment. These Evidence Surfaces Do Not Establish AgentGuard As A Complete AI CISO Platform.

Explore AI CISO EvaluationFor Security Teams

First-Party Advisories
For AI Agent Threats

Review Public AgentGuard Advisories Covering Documented Threats Across Agent Components, Software Supply Chains, Phishing URLs, And Prompt-Injection Payloads.

Browse All AdvisoriesExplore Threat Intelligence
Industry News

Industry Security News

Track the latest industry updates across AI agents, Web3 security, and infrastructure ecosystems.

Start With The Protection
Level You Need

Start With The Official Quickstart, Or Talk To The AgentGuard Team About Integration Depth, Data Handling, And Deployment Requirements.

Free
$0
forever

Local-first protection for individual agents

  • 100 protected actions / month
  • Runtime policy + supply-chain detectors
  • Basic audit timeline
  • Community support
  • AI deep analysis
  • Webhook and approval notifications
  • Custom runtime rules
Recommended
Personal
$15
/month
7-day free trial

Runtime guardrails for personal agent work

  • 500 protected actions / month
  • Runtime policy + supply-chain detectors
  • Detailed action and scan reports
  • AI deep analysis
  • Batch supply-chain scanning
  • Priority support
  • Custom runtime rules
Starter
$99
/month

Team protection for shared agent workflows

  • 10,000 protected actions / month
  • Runtime policy + supply-chain detectors
  • Full runtime and supply-chain reports
  • AI deep analysis
  • Webhook and approval notifications
  • Team management (5 seats)
  • Priority support
Pro
$499
/month

Advanced governance for production agents

  • 100,000 protected actions / month
  • Runtime policy + supply-chain detectors
  • Full reports + redacted audit logs
  • AI deep analysis
  • Webhook and approval notifications
  • Unlimited team seats
  • Custom runtime rules
Enterprise
Custom

Custom controls for regulated agent operations

  • Unlimited protected actions
  • Runtime policy + supply-chain detectors
  • Full reports + redacted audit logs
  • AI deep analysis
  • Webhook and approval notifications
  • Unlimited team seats
  • Custom rules + private deployment + SLA

Frequently
Asked Questions

What is AgentGuard?+

AgentGuard is a security platform for AI agents. It combines runtime protection and component review to help teams identify risks and evaluate observable high-risk actions through supported selected integration paths.

How is AgentGuard different from a prompt filter?+

Prompt filters mainly focus on input or output text. AgentGuard focuses on execution risk: what an agent is about to run, access, or trigger.

What can AgentGuard detect?+

AgentGuard provides two layers of security analysis. Runtime Guard focuses on observable high-risk actions during execution. Deep Scan presents 6 featured public categories; implementation may run additional internal detectors for static risk analysis on agent-related code and components.

What is Daily Patrol?+

Daily Patrol monitors documented security-relevant changes inside a supported OpenClaw workspace. Confirm current availability and environment scope before relying on it.

Does AgentGuard upload my code, prompts, secrets, or file contents?+

In local mode, AgentGuard does not upload full code, prompts, secrets, or file contents. When cloud services are enabled, AgentGuard may upload sanitized action previews and metadata to support review, approval, and audit workflows.

Does AgentGuard support MCP?+

AgentGuard supports MCP server scanning and MCP-related integrations. However, it should not be described as fully monitoring or blocking all third-party MCP runtime calls today.

Which agent platforms does AgentGuard support?+

AgentGuard supports runtime hooks for Claude Code, OpenClaw, and Hermes. For Codex, Gemini, Cursor, and Copilot, support mainly depends on Skill or command-level integrations and may vary by environment.

Add A Security Decision To Your
Agent Workflow

Start With The Official Quickstart, Or Talk To The AgentGuard Team About Integration Depth, Data Handling And Deployment Requirements.