Home

Aguara Wiki

Aguara is a security scanner for AI agent skills and MCP servers. It detects prompt injection, data exfiltration, credential leaks, and supply-chain attacks using static analysis — no API keys, no cloud, no LLM. One binary.

Quick Links

Page	Description
Installation	Install via script, Go, or binary download
Quick Start	Get scanning in 30 seconds
CLI Reference	All commands, flags, and output formats
Detection Rules	148+ rules across 13 categories
Custom Rules	Write your own YAML detection rules
Configuration	.aguara.yml project config
CI Integration	GitHub Actions, GitLab CI, and other pipelines
Go Library API	Embed Aguara in your own Go tools
MCP Discovery	Auto-detect 17 MCP client configurations
Architecture	Internal design and analysis pipeline
Ecosystem	Aguara MCP, Aguara Watch, Oktsec
Contributing	Development setup, adding rules, PR process
FAQ	Common questions and troubleshooting

Project Stats

• 148+ built-in rules across 13 categories
• 5 analysis engines: pattern matching, NLP, toxic-flow, rug-pull, code block awareness
• 17 MCP client auto-discovery (Claude Desktop, Cursor, VS Code, Cline, Windsurf, and more)
• 4 output formats: terminal, JSON, SARIF, Markdown
• 28,000+ skills monitored daily via Aguara Watch

Links

• GitHub Repository
• Releases
• Aguara Watch
• Aguara MCP
• Go Package Docs