Security

PentestCode is an autonomous AI pentesting agent that runs in a terminal, scanning, enumerating and attacking a target while tracking engagement state and coordinating multiple agents in parallel.

• Security

The release repository for Bitwarden's self-hosted deployment, holding the installation scripts and signed container images used to run Bitwarden's password manager on your own infrastructure.

• Security
• Containers

An AI-powered penetration-testing assistant reached by chatting with an agent that can run isolated cloud commands, using an optional less-restricted model provider for security requests standard models tend to refuse.

• Security

An automated vulnerability-research pipeline engine that parses, decompiles and analyzes Windows kernel drivers for exploitable IOCTLs with AI agents, defined declaratively as a resumable, parallel YAML pipeline.

• Security

An open-source AI security agent that runs in a terminal, following runbooks and switching between specialized agents while tracking findings, evidence and reports for authorized application-security and penetration-testing work.

• Security

A standalone server that connects a modified JADX Android decompiler to large language models over MCP, so an assistant can read decompiled app code and follow along during reverse engineering.

• MCP servers
• Security

A security-governance layer for OpenClaw and other agentic AI runtimes that scans capabilities before use, inspects runtime traffic, and exports durable audit evidence.

• AI evals
• Security

A local security-scanning agent for AI-written software that finds issues across code, agent configuration and secrets, then investigates whether each one is real before reporting it as evidence.

• Security

A system that synthesizes correct-by-construction, formally verified code for cryptographic primitives, extracting the result to OCaml, Rust, and Go.

• Security

A local security audit tool for AI API relays and LLM proxies that checks for prompt injection, model substitution, tool-call rewriting, response-stream anomalies and credential-related risk.

• Security

A network scanning and vulnerability assessment tool for wired and Wi-Fi networks, running on a Raspberry Pi or a headless server, with a web dashboard for discovery, threat intelligence, and offensive-security testing.

• Security
• Networking

A Python framework for LLM security research that builds and runs jailbreak attacks against language models from reusable components, then scores how well each attack worked.

• AI evals
• Security

An open-core secrets management platform for teams and AI coding agents, encrypting environment variables and other secrets end to end.

• Security

An agentic AI security tool, developed at Capital One, that reasons forward from attacker-accessible entry points to identify which source-code defects are actually exploitable, then runs a falsification pass before reporting a finding.

• Security

A lightweight, container-free sandbox that wraps commands to block network access by default and restrict filesystem operations by configurable rule, and doubles as a permission manager for AI coding agents.

• CLI tools
• Security

A tool for reproducibly modifying Android A/B OTA update images and re-signing them with custom keys, with subcommands for packing and unpacking several Android image formats.

• Security
• Mobile

An autonomous bug-bounty framework of agents, commands, CLI tools, skills and MCP servers for Claude Code and six other AI coding tools, with hunt loops, an exploit-chain builder and live bug-bounty platform integration.

• Security

An open-source autonomous AI penetration-testing tool that runs specialist agents across web, API, cloud, Active Directory, Kubernetes and CI/CD targets, documenting each finding with the command and raw output that produced it.

• Security

An autonomous penetration-testing agent that drives an OpenAI-compatible LLM gateway against a Kali Linux Docker sandbox, running a structured recon-analysis-exploit-report pipeline through a real-time terminal interface.

• Security

ClawKeeper is host-agnostic safety middleware between an AI agent and its tools, blocking risky shell, Python, browser, filesystem and network calls, redacting sensitive results, and escalating harder decisions to a Watcher that learns new guard patterns.

• AI evals
• Security

Vigolium is a vulnerability scanner that combines a fast, deterministic native scan across 317 modules with an agentic scan mode where an AI plans attacks, selects modules, writes custom extensions and triages results against a codebase.

• Security

rizzo-pii is a small, CPU-friendly model that detects 22 categories of personal data in Italian legal text, including Italian tax and property identifiers, and drives a local, reversible anonymize-then-restore workflow so sensitive documents can still be sent to a cloud LLM.

• Models
• Security

numbat gives visibility into what AI coding agents do on a developer's machine, watching supported desktop, CLI, IDE and gateway agents through local hooks, logs and session files, and can optionally block a risky action before it runs.

• AI evals
• Security

ClawVault is a security layer for OpenClaw-style AI agents that logs and can block what the agent touches, letting a user lock private files, API keys and credentials into a monitored vault and get an instant-messaging alert whenever the agent reaches for them.

• Security