Inkog Documentation
AI Agent Security Platform
Find vulnerabilities in AI agent logic
Prompt injection, tool misuse, infinite loops, missing oversight — found and mapped to compliance frameworks before your agents hit production. Audited on 500+ open-source agents with a 0% false-positive rate.
Quick Start
Scan your agent code in seconds — no install needed:
npx -y @inkog-io/cli scan .Or install permanently and configure your API key:
curl -fsSL https://inkog.io/install.sh | sh
export INKOG_API_KEY=sk_live_your_key_here
inkog scan .Why Inkog?
Agentic Pattern Analysis
Purpose-built for AI agents. Detects token bombing, infinite recursion, and logic flaws that generic SAST tools miss.
Code & No-Code Support
Scan LangChain, CrewAI, AutoGen code alongside n8n, Flowise, and Langflow workflows.
Compliance Ready
Automated mapping to EU AI Act, OWASP LLM Top 10, and NIST AI RMF. Audit reports out of the box.
Supported Frameworks
Supported Frameworks
What Inkog Detects
# Prompt injection vulnerability
def process_query(user_input):
prompt = f"Answer this: {user_input}"
return llm.generate(prompt)# Sanitized input with validation
def process_query(user_input):
sanitized = sanitize_input(user_input)
validate_input(sanitized)
prompt = PROMPT_TEMPLATE.format(query=sanitized)
return llm.generate(prompt)Explore the Docs
Installation Guide
Install Inkog via Docker, Homebrew, or from source. Get scanning in under a minute.
Vulnerability Database
Browse our comprehensive database of AI agent security risks with detection rules and remediation guides.
CLI Reference
Complete reference for all Inkog Verify commands, flags, and configuration options.
EU AI Act Compliance
Map Inkog's capabilities to EU AI Act Article 15 requirements for high-risk AI systems.