security

A minimal native macOS sandbox for running Claude Code or Codex with OS-enforced limits on what the agent can touch.

An open-source reference implementation for autonomous vulnerability discovery and remediation with Claude. It includes Claude Code skills for threat modeling, scanning, triage, patching, plus a harness for running a recon → find → verify → report → patch pipeline.

Open-source security scanner and firewall for MCP servers. It scans configs for common MCP risks, enforces runtime policies on tool calls, and produces audit logs, with CLI commands, policy examples, and a programmatic API.

A secure Model Context Protocol server that gives AI assistants controlled terminal access with risk analysis, sandboxing, logging, filesystem protection, and optional Docker and Git features.

An interactive lab for tracing how AI-assisted engineering workflows can reach repos, CI/CD, credentials, tools, approvals, and proof trails.

Runtime governance and containment for AI agents. Faramesh sits between an agent and its tools to enforce policy checks, approval steps, credential isolation, and tamper-evident audit logs before risky actions execute.

A security product for Claude Code that reviews application architecture inside the IDE via MCP, with contextual assessments, prioritized mitigations, and continuous re-checks as the codebase changes.

A local LLM gateway that redacts secrets before requests reach model providers and restores them locally afterward. It is aimed at agent workflows where prompts and tool calls may contain real credentials.

HEIDES is a deterministic local harness for AI coding agents that maps a codebase, grounds plans against the current graph, and checks proposed patches for conflicts, edge cases, and security issues before changes land.

Sentrint is a security scanner for apps built with LLM platforms like Bolt, Cursor, and v0. It scans repos for secrets, access rules, vulnerable dependencies, and risky code paths, then returns plain-English findings and a copy-paste fix prompt.

A lightweight shell-based guard that blocks destructive commands before an AI coding agent can run them. It targets risky actions like git reset --hard, rm -rf, force-pushes, and destructive cloud/database commands, with local regex rules and a test script.

A CLI for running untrusted MCP servers in sandboxed containers with deny-by-default egress, traffic watching, and host approval workflows. It can also compose servers into agents and expose them over OCI registries.

A harness that runs MCP servers in a disposable Linux sandbox and logs their real startup behavior, including file access and outbound network connections.

An open-source AI agent platform for production workflows, with isolated containers, credential proxying, and support for 100+ business and productivity integrations.

VaultS3 is a self-hosted, S3-compatible object storage server with a built-in dashboard, low RAM usage, and a single-binary deployment. The page highlights features like versioning, WORM, S3 Select, FUSE mount, IAM/OIDC login, search, and event notifications.

A Mac-resident MCP server that lets AI agents access Apple services like Mail, Calendar, iCloud Drive, Voice Memos, Reminders, and Contacts over a Tailscale network with per-token ACLs and audit logging.

HookGuard is a CLI security scanner for AI coding agent configuration files. It looks for malicious hooks, invisible Unicode, credential exfiltration patterns, and prompt-injection text in files like CLAUDE.md, AGENTS.md, Cursor rules, and GitHub Copilot instructions.

HDTP is a protocol specification for agent-to-agent communication on behalf of people, centered on certificates, contact approval, per-caller permissions, and sealed end-to-end calls.

A public timeline and dataset of notable AI-involved hacking and cyber incidents, with RSS support and source links for each entry.

ContextVeil is an open-source tool that redacts enrolled secret values locally before coding agents send text to an LLM. It keeps files and commands usable while replacing exact secret matches with placeholders.

A browser extension that masks personal and sensitive information before you send messages to AI chat tools, then restores the original values in the reply.

A blog post describing Annex, a local-AI security tool that encrypts selected files during agent sessions to keep sensitive data away from local AI tools. The post explains the threat model, encryption flow, and the stack used to build it.

Code Airlock is a wrapper for running coding agents like Claude Code and Codex inside disposable microVMs, with changes committed to git for host-side review.

git-lrc is a Git-based AI code review tool that runs on commit and surfaces risk categories, inline findings, and a summary deck for each review.