Can you trust this AI asset?
Search security intelligence for AI skills, MCP servers, plugins, and a growing range of AI assets. If Manifest doesn't know it, scan it on demand.
Search Manifest
Not seeing the asset you're looking for?
Scan an asset
A file, a folder, or a .zip - whatever holds the asset.
Recently flagged
Latest high-risk findings across the AI supply chain.
home-router-tune
Audit and safely tune a home ISP modem/router across four gated phases — discover the LAN and Wi-Fi neighborhood, map every admin-panel setting, draft a prioritized improvement plan, execute only what the user approves, then write a per-run report plus an append-only history. Reads admin credentials from a local .env (never asked interactively), drives the admin panel via the dev-browser skill, and ships tested vendor playbooks for Sagemcom, Huawei, ZTE, Nokia, Intelbras, and Technicolor gateways used by Claro, Vivo, Oi, and TIM in Brazil; generalizes to any other consumer ISP gateway. Trigger broadly on anything involving the user's home modem, router, Wi-Fi, or ISP-supplied gateway — auditing settings, hardening security (WPS, remote management, default password, firewall, telemetry), troubleshooting flaky or slow Wi-Fi, devices that keep dropping (cameras, smart bulbs, IoT), weak signal in part of the house, neighbor channel interference, post-ISP-swap reviews, or simply logging into 192.168.0.1 / 192.168.1.1 / 192.168.100.1 to change something. Loose phrasings count — "my Wi-Fi is flaky", "the ISP just swapped my modem", "make my home network faster", "minha internet tá ruim", "trocaram meu roteador", "ajustar Wi-Fi", "câmera fica caindo" — and so do casual queries that never use the words "router" or "modem" but describe the symptoms. Skip when the request is app-level networking (Express/Next.js routers, React Router), OS-level laptop network config, cloud VPC/Terraform, nmap/DNS scripting, or a genuine link-layer fault that needs an ISP truck roll rather than a config change.
16-ai-llm-security
LLM and AI application security testing — prompt injection, jailbreak
03-exploit-development
Proof-of-concept development, payload crafting, shellcode analysis, and
run-model-eval
Test a local model server that is already running (Ollama, LM Studio, llama.cpp, MLX, vLLM, SGLang) for speed and practical capability, and dig into anything that looks off.
Latest research
New findings from the Manifold research team.

GitSpawn: A Single Flaw Lets Untrusted Repos Run Code in Claude Code, Codex, Cursor, and Grok
Popular CLI AI coding agents run git commands on startup. A repo you were sent can hijack them to run code on your machine, no clicks required.

OpenAI & Hugging Face: Why the Chain-of-Thought Police Won't Save You
OpenAI saw agents coordinating seven weeks before the Hugging Face breach. Each alert was read alone. One more monitor would not have changed that.

What to Think About curl | bash Now That AI Agents Run It.
AI agents now run curl | bash with your credentials. We scanned 4,003 extensions and 2.7M agent events: the URL tells you nothing. Behavior does.
Manifest registry
Explore the full database
Search, filter, and compare every indexed skill, plugin, MCP server, and editor extension.