Search

A watchtower for AI agents — scan MCP servers and AI-agent tool manifests for poisoned descriptions, data exfiltration, secrets and supply-chain risk. Zero-dependency, CLI + CI.

Model Context Protocol Server for Rebase — exposes schema, DB, document, and user management tools to AI assistants.

MSW (Mock Service Worker) Plugin for ObjectStack Runtime

MCP compatibility adapter for Agent Tool Protocol

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.

[](https://glama.ai/mcp/servers/cuttalo/depscope) 📇 ☁️ 🏠 - Package Intelligence for AI agents. 22 tools across 17 ecosystems (npm/pypi/cargo/go/maven/nuget/rubygems/composer/pub/hex/swift/cocoapods/cpan/hackage/cran/conda/homebrew) — check health, vulnerabilities (OSV + CISA KEV + EPSS), typosquats, malicious flags, alternatives, known bugs, breaking changes, stack compatibility and error-to-fix. 31k+

Supply chain risk scorer for npm, PyPI, Cargo, and Go packages — behavioral signals that can't be faked

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

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

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

Temporary package placeholder for staged publishing

[](https://glama.ai/mcp/servers/elang2/mcp-audit-gateway) 📇 🏠 🍎 🪟 🐧 - Tamper-proof audit trail for AI agent tool calls. Transparent stdio proxy with cryptographic attestation (Ed25519 signing), hash-chained records, YAML policy enforcement (allow/deny/transform), and OpenTelemetry export. Zero server modification required. `npm i @mcp-audit-gateway/core`

Lean & Secure Bot

Scan GitHub repos and profiles for malware before cloning — commit-pinned risk verdicts for agents

Give two agents on two machines the same encrypted folder. P2P, no cloud, only read bytes move.

AI Agent vulnerability scanner — 215 fault types, 88 detection rules, 52 ZDI cases. Scans MCP configs, agent code, framework implementations. 6 CVEs filed.

Audit your package.json dependencies

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

Behavioral trust scores for npm packages, GitHub repos, and businesses.

MCP server that shows what a checkout you did not write tells your agent to do — instruction files, hook commands wired to agent events, MCP servers it declares, and the skills and subagents it ships.