Search

MCP Gateway — one in-path proxy that scores a downstream MCP server and governs every tool call with capability leases. Fuses @gatewarden/score and @gatewarden/govern.

Lean & Secure Bot

Static analyzer and lockfile for AI-agent context files

Lightweight, Chinese-first prompt-injection detection & PII redaction SDK. Inline in agent data flows (LangChain-JS, MCP, any TS runtime). Zero runtime deps, offline, fail-open.

MCP tool for scanning Maven project dependency vulnerabilities

Local first MCP checks for Unicode confusables, mixed scripts, invisible controls, and suspicious identifiers

Protect your AI agents and IDEs from malicious open-source packages.

Scan your machine for known AI-coding-assistant supply-chain backdoor indicators (Claude Code / Cursor / Windsurf extensions, MCP servers) plus risky hook/permission config patterns.

Day-zero supply-chain checker for npm + PyPI, as a local MCP server for AI coding agents. Calls check_package before installs to surface contextual anomalies.

Risk signals for a repository or an npm/PyPI package, with evidence. Never runs the target's code.

Read-only Supabase RLS check: explains in Arabic which tables anyone can read or change. No keys.

MCP server + cross-harness skills for building and operating AI agents safely

Sandboxed shell exec for MCP clients: run untrusted agent commands in a gVisor container.

Sentinel MCP server: agent-native pre-install audit tools backed by the Sentinel proxy (stdio Model Context Protocol server).

Deterministic, non-executing evidence reports for public npm MCP servers.

Check npm packages for typosquats, hallucinated names, install scripts and risky new releases.

Tooling for MCP. Coming soon.

MCP compatibility adapter for Agent Tool Protocol

Audit the hooks, skills, permissions, MCP servers, and configuration your AI coding agents trust.

Sandboxed shell exec for MCP clients: run untrusted agent commands in a gVisor container.