Express Security Scan
Defensive security scan for Express.js applications. Detects missing helmet, unsafe body-parser limits, broken trust-proxy config, weak cookie/session options, missing CSRF, middleware-ordering bugs (auth registered after route), unvalidated res.sendFile, and unsafe eval of request data. Invoke when the user asks to "review", "audit", or "scan" an Express project.
Vercel Ai Sdk Security Scan
Defensive security scan for the Vercel AI SDK (`ai` package). Detects unbounded experimental_attachments, streamText output rendered as HTML, tool execute handlers running shell from args, useChat endpoints without auth, generateObject schemas missing strict mode, and onFinish leaking content to telemetry. Invoke when the user asks to "review", "audit", or "scan" code using the AI SDK.
Express Attack Probe
Authorized self-pentest probe targeting Express.js-specific weaknesses. Tests prototype pollution via merge/clone middlewares, HPP (HTTP parameter pollution) via body-parser, x-forwarded-for spoofing when trust proxy is loose, qs depth/array bombs, sendFile path traversal, and known unsafe middleware patterns. Use when the user asks to "pentest" or "attack-test" their own Express app.
Nestjs Attack Probe
Authorized self-pentest probe targeting NestJS-specific weaknesses. Tests global ValidationPipe gaps, missing @UseGuards on controller methods, default Swagger/OpenAPI exposure, WebSocket Gateway auth bypass, and TypeORM/Prisma raw query injection points discovered via OpenAPI. Use when the user asks to "pentest" their own NestJS app.
Langchain Attack Probe
Authorized red-team probe for applications built on LangChain or LangGraph. Tests PythonREPL/ShellTool exploitation, RAG injection via user-uploaded documents, output-parser injection, callback-handler leakage, and SQL agent privilege checks against your own endpoint discovered via env/entrypoint. Use when the user asks to "red-team" or "attack-test" their LangChain app.
Mcp Server Security Scan
Defensive security scan for Model Context Protocol (MCP) servers built with the official @modelcontextprotocol/sdk or mcp Python SDK. Detects tools exposing filesystem/shell with user-controlled paths, HTTP/SSE transports without authentication, unvalidated tool arguments, sensitive data leaking via resource URIs, and unbounded tool output that floods context. Invoke when the user asks to "review…
Fastapi Security Scan
Defensive security scan for FastAPI applications. Detects missing Depends/Security guards, Pydantic validation bypasses, permissive CORS, unverified JWTs, raw SQL string interpolation, and unsafe file responses. Invoke when the user asks to "review", "audit", or "scan" a FastAPI project, or when editing routers and dependencies.
Mcp Server Attack Probe
Authorized red-team probe for Model Context Protocol (MCP) servers. Tests tool path traversal, command injection in shell-like tools, SSRF via fetch tools, HTTP/SSE transport authentication, DNS rebinding via Origin header, and resource URI traversal. Use when the user asks to "red-team" or "attack-test" their own MCP server.
Openai Sdk Security Scan
Defensive security scan for code using the OpenAI SDK (openai Python, openai Node). Detects API keys in client bundles, function-calling handlers without argument validation, missing structured-output schema, untrusted attachments fed to vision/Assistants, model output rendered as HTML, and verbose logging of full prompts. Invoke when the user asks to "review", "audit", or "scan" code that calls…
Spring Boot Attack Probe
Authorized self-pentest probe targeting Spring Boot-specific weaknesses. Tests Actuator endpoint exposure (/env, /heapdump, /loggers), Whitelabel error page info disclosure, h2-console exposure, Spring Security permitAll gaps, and SpEL/parameter-binding pitfalls. Use when the user asks to "pentest" their own Spring Boot app.
Vercel Ai Sdk Attack Probe
Authorized red-team probe for applications built on the Vercel AI SDK (`ai` package). Tests tool execute exploitation via crafted user messages, useChat endpoint authentication, attachment limits, and streamText/dangerouslySetInnerHTML XSS via injected markdown. Use when the user asks to "red-team" or "attack-test" their AI SDK app.
Spring Boot Security Scan
Defensive security scan for Spring Boot applications using Spring Security. Detects permitAll on sensitive routes, disabled CSRF on stateful endpoints, wildcard CORS with credentials, missing @PreAuthorize, JdbcTemplate string concatenation, Jackson default typing, exposed actuators, and weak BCrypt strength. Invoke when the user asks to "review", "audit", or "scan" a Spring Boot project.
Fastapi Attack Probe
Authorized self-pentest probe targeting FastAPI-specific weaknesses. Tests /docs and /redoc auth, OpenAPI schema enumeration, Pydantic boundary bypass via extra fields, missing Depends/Security on routes, JWT alg confusion, and unsafe file responses. Use when the user asks to "pentest" their own FastAPI app.
Nextjs Security Scan
Defensive security scan for Next.js App Router projects. Detects environment-variable leakage to the browser bundle, unauthenticated server actions, missing middleware auth, unsafe CORS on route handlers, and trust-boundary violations in revalidatePath/revalidateTag. Invoke when the user asks to "review", "audit", or "scan" a Next.js codebase, or when editing files under app/ or middleware.ts.
Anthropic Sdk Security Scan
Defensive security scan for code using the Anthropic SDK (@anthropic-ai/sdk, anthropic Python SDK). Detects prompt-injection vectors via untrusted attachments and tool results, system prompts containing secrets, prompt-cache breakpoints over user data, missing tool input validation, and unsafe execution of model output. Invoke when the user asks to "review", "audit", or "scan" code that calls mes…
Langchain Security Scan
Defensive security scan for LangChain / LangGraph applications. Detects unsafe agents (PythonREPLTool, ShellTool), retriever trust-boundary violations, output parser injection, callback handlers leaking secrets to logs, and missing tool input validation. Invoke when the user asks to "review", "audit", or "scan" code using langchain, langgraph, or related extensions.
Django Security Scan
Defensive security scan for Django and Django REST Framework projects. Detects DEBUG=True in production, wildcard ALLOWED_HOSTS, SECRET_KEY in source, missing CSRF, raw ORM queries with string formatting, mark_safe on user input, AllowAny on mutating DRF views, and ModelSerializer fields="__all__" leaking sensitive fields. Invoke when the user asks to "review", "audit", or "scan" a Django project.
Anthropic Sdk Attack Probe
Authorized red-team probe for applications built on the Anthropic SDK. Tests Claude-specific quirks - XML/tag confusion attacks, tool_use schema bypass, prompt-cache cross-tenant probing, and assistant-prefill manipulation - against your own chat or agent endpoint discovered via env/entrypoint. Use when the user asks to "red-team" or "attack-test" their Claude-based app.
Nextjs Attack Probe
Authorized self-pentest probe targeting Next.js App Router-specific weaknesses. Tests NEXT_PUBLIC_* secret leakage in client bundles, server-action invocation without auth, ISR/cache poisoning via Vary mishandling, route-handler CORS misconfig, image proxy SSRF, and middleware matcher gaps. Use when the user asks to "pentest" their own Next.js app.
Openai Sdk Attack Probe
Authorized red-team probe for applications built on the OpenAI SDK. Tests function-calling schema bypass, structured-output schema escape, Assistants thread cross-tenant access, and vision payload injection against your own endpoint discovered via env/entrypoint. Use when the user asks to "red-team" or "attack-test" their OpenAI-based app.
Django Attack Probe
Authorized self-pentest probe targeting Django and DRF-specific weaknesses. Tests DEBUG-page leak, ALLOWED_HOSTS bypass via Host header, default admin path enumeration, ORM raw() injection, mass-assignment via ModelSerializer, and DRF AllowAny on mutating endpoints. Use when the user asks to "pentest" their own Django app.
Nestjs Security Scan
Defensive security scan for NestJS applications. Detects missing global ValidationPipe with whitelist, controllers without @UseGuards, DTOs without class-validator decorators, permissive CORS, missing helmet, exception filters leaking stack traces, TypeORM raw queries with template literals, and WebSocket gateways without auth. Invoke when the user asks to "review", "audit", or "scan" a NestJS pr…
Openapi Spec Security Scan
Defensive security scan for OpenAPI 3.x specifications (openapi.yaml / openapi.json). Detects missing global security, unprotected mutating operations, HTTP-only servers, loose schemas (additionalProperties true, missing required), wildcard CORS, missing 401/403 responses, PII in examples, and weakly-typed parameters. Invoke when the user asks to "review", "audit", or "scan" an OpenAPI / Swagger…
Prompt Injection Probe
Authorized red-team probe for prompt-injection vulnerabilities in your own LLM application. Submits a battery of direct and indirect injection payloads against a chat/agent endpoint discovered via env/entrypoint, and scores how often the system prompt, tool boundary, or output policy is broken. Use when the user asks to "test prompt injection on" or "red-team" their own LLM app. Pair with framewo…
Webapp Pentest Checklist
Authorized self-pentest checklist for web applications you own. Walks the OWASP Web/API Top 10 against a locally-running target, discovering the base URL via env or entrypoint files (never hardcoded). Use when the user asks to "pentest", "attack", "probe", or "test the security of" their own running web app and the framework is unknown or mixed. For framework-specific deeper checks, use express/d…