Security scanner for AI agent capability manifests — detects prompt injection, permission creep, and data exfiltration

Security scanner for AI agent capability manifests — detects prompt injection, permission creep, and data exfiltration

Dependency-free heuristic scanner that detects prompt-injection, jailbreak, and data-exfiltration patterns in untrusted text before it reaches an LLM.

Supply-chain firewall for AI agent skills, MCP servers, and rules files. Scans for prompt injection, data exfiltration, and tool poisoning before you install — and emits an agent-BOM + compliance evidence as a byproduct.

Security moat for AI agents. Runtime protection against prompt injection, tool misuse, and data exfiltration.

Security scanner for AI agent skills and prompts - detects prompt injection, jailbreaks, and data exfiltration

Context-aware security scanner for AI agent skills, MCP servers, and tool configs. 95+ rules with false-positive reduction. Detects prompt injection, credential leaks, CORS misconfig, SSRF, tool poisoning, and data exfiltration.

This repository contains 2 related npm packages concerned with hardening markdown against [data exfiltration attacks through LLM prompt-injection](https://vercel.com/blog/building-secure-ai-agents#exfiltration-through-model-output).

Safety checks and pattern databases for AI coding agents. Intercepts tool calls to block destructive commands, credential exposure, prompt injection, and data exfiltration.

Security scanner for AI agent skill packages — detect prompt injection, data exfiltration, command injection, and more

Citadel Guard plugin for OpenClaw - AI security protection against prompt injection, data exfiltration, and more

MCP server that scans other MCP servers for prompt injection, data exfiltration, and privilege escalation. Add to your .mcp.json and let your AI agent audit its own tools.

Security scanner for OpenClaw skills - detect malicious patterns, data exfiltration, and supply chain risks

Pre-clone static security scanner for GitHub repositories and local codebases. Detects malicious lifecycle scripts, credential theft, obfuscated payloads, trojan source, typosquatting, data exfiltration, and lockfile integrity issues.

Dependency-free static analyzer that audits LLM agent tool/function definitions (OpenAI, Anthropic, MCP) for excessive agency, code-execution & SSRF sinks, secret access, and data-exfiltration surface — mapped to the OWASP 2025 LLM Top-10. Runs in CI.

The CDK Construct Library for AWS::Route53Resolver

Security scanner for AI agent skills/plugins. Detects prompt injection, data exfiltration, excessive permissions, and supply chain attacks.

Security scanner for PostHog's agentic flows

SAFE test package: exfiltration-intent patterns with dummy data; localhost only; optional CLI.

Pre-install supply-chain poison scanner for DeepSeek Harness plugins: AST analysis (NodeSecure JS-X-Ray) + deobfuscation decoder + regex heuristics to catch credential exfiltration, dynamic code execution, obfuscated imports, and install-time scripts.