BotGuard CLI

Security scanning for AI agents from your terminal. Tests for prompt injection, jailbreaks, data extraction, and 70+ attack scenarios aligned to OWASP LLM Top 10.

Install

npm install -g botguard-cli
# or run directly
npx botguard-cli scan

Quick Start

# Run a security scan
botguard scan --api-key YOUR_KEY --endpoint https://my-agent.com/chat --description "My chatbot"

# Scan with a system prompt file
botguard scan --api-key YOUR_KEY --system-prompt ./prompts/system.md --fail-threshold 80

# Check remaining credits
botguard credits --api-key YOUR_KEY

# Generate a config file
botguard init

Config File

Create a .botguard.yml to avoid passing flags every time:

apiKey: ${BOTGUARD_API_KEY}
scan:
  endpoint: https://my-agent.com/chat
  description: "Customer support chatbot"
  categories: [jailbreak, prompt_injection, data_extraction]
  failThreshold: 80
  format: table

Then just run:

botguard scan

Options

| Flag | Description | Default | |---|---|---| | --api-key | API key (or BOTGUARD_API_KEY env var) | — | | --endpoint | Agent API endpoint URL | — | | --description | Agent description | — | | --system-prompt | System prompt (text or file path) | — | | --mode | sync or async | sync | | --categories | Comma-separated attack categories | all | | --attack-count | Number of attacks | — | | --fail-threshold | Exit 1 if score below this (0-100) | 0 | | --format | table, json, or summary | table | | --config | Path to .botguard.yml | auto-detect |

CI/CD Usage

The CLI returns exit code 1 when the score is below --fail-threshold, making it CI-friendly:

botguard scan --fail-threshold 80 --format summary

For GitHub Actions, use botguardai/security-scan which wraps this CLI with PR comments and check status.

Get Your API Key

1. Sign up at botguard.dev
2. Go to Account > API Keys
3. Create a CI/CD key

Related

• BotGuard — Automated red-teaming & real-time firewall for AI agents
• GitHub Action — CI/CD security scanning
• Attack Library — 229+ open-source LLM attack templates

License

MIT