security-testing

CLI-first authorization regression testing with bounded browser exploration and deterministic paired replay.

Defensive security checks a QA team can run in CI, against OWASP Juice Shop running locally in Docker.

Security-native AI IDE for evidence-driven code review, authorized pentesting, verified fixes, and MCP integrations.

Simple express typescript api boilerplate

Replay-verified MCP security testing and deterministic policy hardening for tool-using AI agents

Full-stack survey platform: Next.js 15, PostgreSQL, real-time analytics. Security-tested (OWASP ZAP), rate-limited, privacy-first architecture.

MCP server that lets LLMs drive HTTPToolkit for automated API security testing — replay, mutation, and audit on captured mobile, desktop, and CLI traffic

Functional + security testing for MCP servers — deterministic injection probes, one YAML, one exit code

assistant & TUI with local model support: llama.cpp, verified token-saving eco mode

Cypress + Playwright healthcare test automation with AI: self-healing locators, visual/a11y checks, security testing.

RAG エンドポイントへのプロンプトインジェクション耐性診断 CLI — npx pijack test <url>

Three VS Code extensions for security and systems engineering, plus an extension pack

Deliberately-vulnerable TypeScript e-commerce agent: security-scanner validation fixture (do not use)

Proprietary authorized web load / DDoS-simulation testing platform. Testing is restricted to targets the operator owns or is explicitly authorized to test.

QA-specialized fork of pi-coding-agent — 7 testing subagents (planner/web/api/visual/perf/security/redteam) + MCP bridge (Playwright, Chrome DevTools, axe, nuclei). All subagents on claude-opus-4-7.

AI penetration testing platform with autonomous security agents

Security testing that runs inside the coding agent you already use. Source-available, not open source.

End-to-end test automation harness and CI/CD security validation pipeline for OWASP Juice Shop.

Replay captured web requests across sessions and see which routes remain open to the wrong user.

LLM security red teaming — Claude Code plugin + MCP server with 80+ attack plugins and 25+ mutation strategies