Security Guards | CrabShack Docs

📄️ Prompt Injection

Detects and neutralizes attempts to hijack your AI agent's instructions.

Prevents your agent's instructions from being extracted or leaked.

Controls which tools your AI agent can use and monitors for suspicious usage patterns.

Restricts file access to authorized areas and protects sensitive files.

Detects and redacts API keys, passwords, tokens, and personal information in conversations.

Detects commands that attempt to gain elevated system access or escape containers.

Enforces message size limits and rate limiting to prevent abuse.

Scans AI-generated responses for dangerous code patterns and security vulnerabilities.

Protects against poisoned memories and enforces session size limits.

Controls which external tool servers your agent can connect to and inspects their responses.

Controls where your agent can make network requests and blocks access to internal resources.

Ensures agent skills are authentic and haven't been tampered with.

Monitors critical files for unauthorized changes, deletions, or unexpected additions.

Ensures all messages come from authorized channels and verified users.

Complete security event logging with configurable retention and real-time alert delivery.

Notices when an agent starts behaving unusually compared to its own normal pattern.

Lets you tell CrabShield about specific, known-safe behavior so it stops flagging it.

Checks your own CrabShield settings for gaps, so protection isn't weakened by accident.

Scans documents you upload to your Knowledge Base for hidden manipulation attempts before they reach your agents.