Code-only MCP servers — agent-tools.cloud

MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.

Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.

attack-surface-mcp-server

github.com/cyanheads/attack-surface-mcp-server
active

Passive external attack-surface mapping: CT subdomains, DNS, TLS, HTTP posture, RDAP/WHOIS, Shodan via MCP. STDIO or Streamable HTTP.

npm Apache-2.0 ★ 3
entity-attack-surface-mcp

github.com/apifyforge/entity-attack-surface-mcp
updated this year

Entity attack surface mapping for any domain — via the Model Context Protocol.

MIT
@cyanheads/attack-surface-mcp-server

github.com/cyanheads/attack-surface-mcp-server
active

Passive external attack-surface mapping: CT subdomains, DNS, TLS, HTTP posture, RDAP/WHOIS, Shodan via MCP. STDIO or Streamable HTTP.

npm Apache-2.0
hopper-recon

github.com/iksnerd/hopper-recon
active

Self-hosted attack-surface recon — point it at a domain you own and map subdomains, DNS, TLS, HTTP, CDN, and historical URLs in one dashboard. The same tools run over MCP, so AI agents (Claude Code, Cline) can drive recon too.

MIT ★ 2
AI-Surface

github.com/apisec-inc/AI-Surface
active

Find and govern AI attack surfaces in application code, at PR time and inside your AI coding tool (MCP server, Claude Code hook). Free, OSS, runs offline.

MIT ★ 144
pentest-suite

github.com/laurentvv/pentest-suite
active

Lightweight attack-surface discovery and vulnerability intelligence toolkit in pure Python stdlib (zero dependencies), with CISA KEV/cve.now feeds, OODA loop orchestration, and MCP server.

MIT
bevigil-mcp-server

github.com/santhosh-005/bevigil-mcp-server
active

MCP server for the BeVigil OSINT API — mobile app attack surface discovery (hosts, subdomains, S3 buckets, wordlists) for AI agents

npm MIT
@vexralabs/zerodom

github.com/devhusnainai/zerodom
active

The deterministic attack-surface perception layer for AI security agents. Maps every actionable element of a live, logged-in Chrome page to a unique id; no LLM in the parse (TypeScript port).

npm Apache-2.0
outrider-recon

github.com/Ap6pack/outrider-recon
active

Claude-native, governed offensive-OSINT & external attack-surface platform for authorized red-team and bug-bounty. Multi-agent recon across web, identity, cloud, secrets & breach intel under deterministic scope/evidence/approval controls, with human-reviewed findings for hand-off to pen-testing. Not an exploitation framework.

NOASSERTION ★ 15
blitzstrike

github.com/shinthink/blitzstrike
active

Blitz Strike — a universal MCP security-audit toolbelt. BLITZ sweeps the attack surface, EAGLE-EYE traces source-to-sink, STRIKE verifies live. 57 attack chains, 130-tool catalog, intelligence data layer. One server, every agent.

npm MIT
ExternalAttacker-MCP

github.com/MorDavid/ExternalAttacker-MCP
stale

A modular external attack surface mapping tool integrating tools for automated reconnaissance and bug bounty workflows.

★ 78
digital-infrastructure-exposure-mcp

github.com/apifyforge/digital-infrastructure-exposure-mcp
updated this year

**Digital infrastructure exposure reconnaissance** delivered as an MCP server — query it from Claude, Cursor, or any AI agent to assess any domain's attack surface in seconds.

MIT
falconeye

github.com/osintph/falconeye
active

Free self-hosted OSINT investigator's toolkit with an MCP server for Claude Code and Claude Desktop. 19 tools in one interface: crypto tracing, phishing fingerprinting and hunting, attack surface, route map, domain intel, Telegram OSINT, IP reputation, email header forensics, username enumeration, HIBP breach checks, ransomware tracking. AGPL-3.0.

AGPL-3.0 ★ 16
ISEEYOU

github.com/felixpalingan/ISEEYOU
active

Interactive OSINT Lead-Chaining Caseboard, Target Dossier Workbench, Wayback Machine Crawler & BYOK AI Detective

MIT
sn1per-mcp

github.com/1N3/sn1per-mcp
active

Sn1per MCP Server: drive your self-hosted Sn1per from any MCP client. Authorized testing only.

Apache-2.0 ★ 3
srcradar-mcp-server

github.com/usdagfhjkda/srcradar-mcp-server
active

面向 agent 的持久化业务级攻击面知识库。把 srcradar 的"业务名 → 法律实体图谱 → 主动测绘资产(小程序/公众号 + Web 范围 → 子域名/端口/指纹) → 每日增量 diff"通过 MCP 协议暴露出来,让任何 agent 无需重新扫描即可直接拿到已知资产清单与最新变动 —— 跳过信息收集,从已识别的目标开始

Apache-2.0
OpenOSINT

github.com/OpenOSINT/OpenOSINT
active

Open-source OSINT agent: 21 public-source tools behind natural-language chat, an MCP server and a Claude Code plugin. Passive by default, no keys to start.

pypi MIT ★ 1,936
qualys-cli-mcp
active

Model Context Protocol (MCP) server giving any LLM secure, audited access to the Qualys Cloud Platform (VM, PM, WAS, TotalCloud, Container Security, CSAM, and more).

pypi MIT
zerodom

github.com/DevHusnainAi/zerodom
active

Point an app at two logged-in users and it finds the broken access control (IDOR). Deterministic browser perception + cross-identity authz testing for AI agents — drives your real Chrome over MCP or runs headless. Open-core, local, no account.

Apache-2.0 ★ 3