Complete Feature Set

Everything you need to secure your development workflow, from command validation to CVE scanning

Command Risk Guard

Automatically blocks dangerous operations before they execute

Vectra Guard analyzes every command for potential risks using 200+ patterns. It prevents root deletion, privilege escalation, and other destructive operations.

Command Risk Guard

# Dangerous command blocked automatically
vg exec -- rm -rf /

❌ CRITICAL: Command blocked for safety
• Pattern: DANGEROUS_DELETE_ROOT
• Risk Level: CRITICAL
• Blocked to protect your system

Smart Sandboxing

Isolate risky commands with intelligent caching for 10x speedup

Run untrusted code in secure Docker containers with mounted dependency caches. First run builds cache, subsequent runs are blazing fast.

Smart Sandboxing

# Always-on sandbox with caching (default)
sandbox:
  enabled: true
  mode: always
  enable_cache: true

# Result: 10x faster installs!
vg exec -- npm install
# First run: 12.3s
# Cached runs: 1.2s ⚡

CVE Scanning

Detect vulnerable dependencies before installation

Local CVE database powered by OSV. Scan package manifests and lockfiles for known vulnerabilities across npm, pip, go, cargo, and more.

CVE Scanning

# Sync CVE database
vg cve sync --path .

# Scan dependencies
vg cve scan --path .

⚠️  lodash@4.17.20 (npm)
├─ CVE-2020-28500 (Moderate)
│  Regular Expression Denial of Service
└─ Fix: Upgrade to lodash@4.17.21

Session Auditing

Complete audit trail of all agent and developer activities

Track every command, file operation, and risk decision. Perfect for compliance, debugging, and understanding AI agent behavior.

Session Auditing

# Start tracked session
SESSION=$(vg session start --agent "cursor-ai")
export VECTRAGUARD_SESSION_ID=$SESSION

# All commands auto-tracked
npm install
git commit -m "feat"

# View session
vg session show $SESSION

10x Faster Installs

Cache-mounted dependencies eliminate repeated downloads

Mount host package caches into sandbox containers. npm, pip, cargo, go, and more automatically reuse cached packages for massive speedup.

10x Faster Installs

# Automatic cache mounting
~/.npm      → container /.npm
~/.cargo    → container /.cargo
~/.cache/pip → container /.cache/pip

Result:
• First install: 12.3s
• Cached install: 1.2s
• 10x faster! ⚡

Explainable Risk

Understand why commands are flagged as risky

Human-friendly explanations for every security decision. Learn security best practices as you work.

Explainable Risk

# Explain specific risks
vg explain risky-script.sh

⚠️  RISK: Detected 'curl | sh'
• Pattern: PIPE_TO_SHELL
• Risk: Remote code execution
• Recommendation: Download first, review, then execute

Trust Store

Approve commands once, run them instantly forever

Build a library of trusted commands that skip sandbox. Perfect for common operations you run daily.

Trust Store

# Trust a command
vg trust add "npm test" --note "Safe test"

# Runs instantly on host (no sandbox)
vg exec -- npm test
✅ 42 tests passed (0.8s)

# Manage trust store
vg trust list
vg trust remove "old-command"

Real-Time Protection

Analyze commands as they execute

Zero-overhead analysis for most commands. Intelligent routing between host and sandbox based on risk.

Real-Time Protection

# Auto mode: smart routing
sandbox:
  mode: auto

# Safe commands → host (instant)
vg exec -- echo "hello"

# Risky commands → sandbox
vg exec -- curl remote.com | sh

Agent-Friendly

Seamless integration with AI coding tools

One command seeds instructions for Cursor, Copilot, and other AI assistants. They learn to use Vectra Guard automatically.

Agent-Friendly

# Seed agent instructions
vg seed agents --target . --targets "agents,cursor"

✅ Created .cursorrules
✅ Created .agents/AGENTS.md

Agents now know:
• When to scan for CVEs
• How to validate risks
• When to use sandbox

Security Without Compromise

Everything you need to protect your development workflow — from AI agents to production deployments.

Command Risk Guard

Blocks dangerous operations automatically before they hit your system.

Smart Sandboxing

Isolates risky actions in secure environments with intelligent caching.

CVE Scanning

Flags packages with known vulnerabilities before installation.

Session Auditing

Track what ran, who ran it, and why — perfect for compliance.

10x Faster Installs

Cache-mounted dependencies make sandboxed installs blazing fast.

Explainable Risk

Human-friendly explanations for why something is risky.

Trust Store

Trust common commands once to skip sandbox on repeated runs.

Real-Time Protection

Analyze and protect commands as they're executed.

Agent-Friendly

Works seamlessly with Cursor, VS Code, Copilot workflows.

30+

Protected Directories

200+

Risk Patterns Detected

Supported Ecosystems

10x

Faster with Cache

Ready to get started?

Install Vectra Guard in 30 seconds and secure your entire workflow

Get Started View on GitHub