"An overview of end-to-end testing" matching MCP servers:

Supply chain risk scoring for npm, PyPI, and GitHub repos

Enables agents to prove and verify that a real, unique human performed an action, issuing machine-readable signed credentials with public verification.

An MCP server that sends plain-text Gmail messages only to a predefined allowlist, using generated recipient IDs instead of email addresses to enforce security.

Provides access to over 40 industry-standard penetration testing tools, including Nmap, SQLMap, and Metasploit, within an isolated Kali Linux Docker container. It enables security professionals to perform comprehensive network reconnaissance, web application testing, and vulnerability research through natural language commands.

MCP server with intentionally malicious capabilities for testing Cisco AI Defense supply chain scanning.

Provides penetration testing tools including nmap, nikto, sqlmap, wpscan, and exploit database searches for educational and authorized security testing purposes using Kali Linux tools.

Enables AI agents to conduct authorized penetration tests by providing scope-enforced access to Metasploit Framework's RPC, with tools for reconnaissance, exploitation, session management, and post-exploitation.

Enables end-to-end testing of OAuth authorization for an MCP server, exposing scoped tools like whoami and echo over streamable HTTP.

Enables MCP clients to exchange an MCP-audience token for a Microsoft Graph token and call Graph as the signed-in user, with tools for identity verification and recent mail access.

Lets AI assistants share HTML they generate as a private, end-to-end encrypted link, and update the page later at the same link. The page is encrypted locally before upload, the key stays in the link, and no account is needed.

Enables an AI agent to send files end-to-end encrypted, encrypting on the user's computer and keeping the download link key, password, and login away from the model.

Lets an AI agent request API keys and passwords from a human through an end-to-end encrypted one-time link and pass them to exactly one command or file, without the value ever entering the agent's context. Can also send secrets as self-destructing links.

Open end-to-end encrypted messaging protocol and MCP server (a4a-mcp) that gives AI agents a cryptographic address (apub1...) to exchange messages, tasks, and signed agreements directly across NAT without port forwarding.

Automated security red-team for any MCP server that scans manifests against OWASP LLM Top 10 and MCP-specific risks, returning a 0-100 hardening score and HMAC-signed report.

Implements a secure MCP server with API Key and JWT authentication, providing tools like echo, login, secure_action, and admin_action. Includes MCP Inspector integration for testing and debugging.

Lets AI assistants manage software license keys end to end, including issuing, suspending, revoking, freeing machines and IPs, blocking leaked keys, looking up customers, building automations, reading validation stats, and adding license checks to code.

Enables an AI agent to pair once with your own phone or computer and then send files and text to it, or receive files you send back, all end-to-end encrypted with nothing arriving unasked. The agent can reach only devices you paired and approved.

Cryptographic receipts for AI agent decisions and game outcomes. Notarize decisions, anchor commitments on Base mainnet, and verify receipts end to end - hash content, submit anchors, check status. Ships 17 tools with both a hosted remote endpoint and a local npm package.

Enables AI assistants and other MCP clients to search, retrieve, and browse end-to-end encrypted contacts locally through read-only tools, decrypting data only on the user's device.

Enables AI agents to search, fetch, and interactively browse web pages through a fail-closed Tor/VPN-only tunnel, rotate identities, and exchange end-to-end encrypted messages with other agents via MCP.

Enables AI agents to list and read Kafka topics and to propose schema-checked message writes that are dry-run by default and require explicit human approval, with every action bounded by the broker's own ACLs for the caller's principal and traced end to end.

MCP server for VoidSend enabling AI agents to send and receive end-to-end encrypted messages.

Publishes AAIF Agent Cards and bridges A2A and OASF to AAIF, with built-in EU AI Act compliance.

Enables AI agents to exchange end-to-end encrypted messages, negotiate availability and plans, and share notes in project spaces through a local node exposing 13 MCP tools and an encrypted relay that only stores ciphertext.

Provides secure, ephemeral secret management for AI agents via MCP tools that self-destruct after use or session end.

A thin MCP server wrapper that automatically audits all tool calls via Attestive's AuditClient, logging decisions and enabling verification of the hash chain. It provides an example server with audited tools and demonstrates end-to-end recording and verification through the MCP protocol.

Enables coding agents on different machines to exchange end-to-end encrypted messages through a shared, self-hosted relay, exposing inboxes, room history, and notifications to Claude Code, Codex, or any MCP client. Peers' messages arrive automatically as clearly framed untrusted data, so collaborators can coordinate without granting tool, file, or command access.

Enables Claude to list Iceberg tables and run read-only DuckDB SQL queries, including time travel by snapshot, with per-caller token authentication and audit logging.

Enables MCP slots to store, share, and read secrets and configuration files through OpenBao, with every value sealed end to end so only ciphertext crosses the broker, transports, and logs. Access is decided by the broker's policy through audiences, keeping read, write, and share rights enforceable without exposing content to unauthorized callers.

Provides MCP tools for interacting with the Technocore agent protocol, including signed messaging, nonce management, DID verification, end-to-end encrypted communication, and untrusted data fencing.