AI

Satori AI lets you describe what you want to test in plain language and generates a ready-to-run playbook for you. Whether you are a developer, a pentester, a QA engineer, or a CTO, you can tell it what to check and it will produce the right playbook for your role.

Watch it in action:

Watch on YouTube

Requirements

Satori AI requires the Claude Code CLI to be installed on your machine:

npm install -g @anthropic-ai/claude-code

You also need a valid Anthropic API key configured for Claude Code. See the Claude Code documentation for setup instructions.

Usage

Satori AI works in two modes: interactive and one-shot.

Interactive Mode

Launch an interactive session where you can have a conversation about what you want to test. The AI will generate playbooks, explain them, and refine them based on your feedback:

satori ai

This opens a conversation where you can say things like:

• "Test my Python app for SQL injection"
• "Check if our API returns the right headers for SOC2 compliance"
• "Scan this repo for hardcoded secrets"
• "Create a monitor that checks our website every hour"

The AI knows the full Satori playbook language, all public playbooks, and all assertion types. It will write a .yml file and give you the command to run it.

One-Shot Mode

If you already know what you want, pass it directly as a prompt:

satori ai "test my Django app for common security issues"

The AI will generate the playbook and provide the execution command without entering an interactive session.

Playbooks for Every Role

Different people in an organization need different tests. Satori AI understands this and generates playbooks appropriate for each role:

For the CEO or compliance officer — verify SOC2 controls, check TLS configuration, security headers, and exposed ports. The result is a report that states which controls passed and which did not, with their severities:

satori ai "verify SOC2 compliance for our website https://example.com"

For the CTO — run a web security audit that reviews server configuration, checks for common misconfigurations, and validates that infrastructure meets security standards:

satori ai "run a security audit on our web server at https://example.com"

For the development lead — scan Docker images for known vulnerabilities before allowing them into production, or check that new code does not introduce regressions:

satori ai "scan our Docker image myapp:latest for vulnerabilities"

For the pentester — launch network scans from ephemeral containers, detect open ports, enumerate services, and scan for known vulnerabilities at scale:

satori ai "perform a network scan on 10.0.0.0/24 looking for exposed services"

For CI engineers — generate playbooks that run on every push to detect issues before code is merged:

satori ai "create a CI playbook that runs semgrep and trufflehog on every push"

For QA engineers — create tests that verify application behavior matches expectations, track regressions, and keep a record of what was tested:

satori ai "test that our API at https://api.example.com/health returns status 200"

What the AI Knows

When you launch satori ai, it has access to:

• The complete Satori playbook syntax, including all assertion types, input formats, and settings
• All public playbooks in the Satori playbook library
• The playbook validation rules from the playbook-validator
• All CLI commands and their options

It clones the Satori repositories to ~/.satori/ on first use to keep its knowledge base up to date.

Examples

Generate a playbook and run it immediately:

satori ai "check if port 443 is open on satori.ci and verify the TLS certificate is valid"

Generate a monitoring playbook that runs on a schedule:

satori ai "monitor our API every 30 minutes and alert on Slack if it returns anything other than 200"

Generate a comprehensive security check for a repository:

satori ai "create a full security playbook for a Python project: lint, secrets, SAST, and dependency audit"

---

If you need any help, please reach out to us on Discord or via Email