Security

Read, write and scan process memory, Cheat Engine style, on Windows, Linux and macOS.

The deterministic attack-surface perception layer for AI security agents

78 pay-per-call tools over MCP: LLM completions, on-chain reads on five EVM chains, web and network checks, sanctions screening, media (image, speech, transcription) and utilities. Listing the tools is free; each call is paid in USDC over x402, with no API key and no account.

Model Context Protocol server for MoveMentors, the booking platform for independent movement teachers and small studios. Search classes, book and reschedule for the signed-in client, and let a mentor or studio owner manage classes, schedules and bookings. OAuth 2.1 with PKCE; see /auth.md.

Remote MCP server. Lets an AI client start a scan, poll its status and read back ranked findings and fixes without leaving the editor. OAuth-protected; the authorization metadata is published on the server origin.

Dead code, security, secrets detection and code quality for Python, TypeScript, Go.

Secret management and multi-chain transaction signing for AI agents via MCP

Deterministic, no-LLM security scanner for MCP servers, plus an inline proxy gate.

Local-first AI identity for MCP coding tools. Lessons, decisions, and context you control.

On-demand Obsidian vault access for AI coding assistants — 17 tools, 5 resources, 4 prompts.

Run commands with API keys injected per call — values never enter model context.

AI agent identity, signed credentials, delegated authority and governance evidence.

Offline security scanning for AI-generated code: seven open-source scanners in one MCP server.

Governed Keycloak + authentik identity ops: users, events, clients, MFA, RCA. 29 tools.

Scan, fix and verify a domain's DNS over MCP: SPF, DMARC, DKIM, mail routing, DNS health, blacklists, multi-region propagation, lookalike-domain checks and domain/SSL expiry. Every record is generated and validated by a deterministic engine, never a language model.

Remote MCP server exposing SafePrompt prompt-injection / jailbreak / code-injection detection as validate_prompt and validate_prompts tools over Streamable HTTP at https://mcp.safeprompt.dev/mcp. Also on npm as @safeprompt.dev/mcp and the official MCP Registry (dev.safeprompt/mcp). Bring your own A…

One MCP endpoint for all your servers: a few meta-tools, Keychain secrets, sandbox, audit, OAuth.

Local operator tools over your own sessions, knowledge, files, shell and devices.

Servidor MCP (Streamable HTTP, JSON-RPC por POST) para que el asistente de un emisor consulte SU cuenta de Facturel: comprobantes emitidos con su CAE, clientes, artículos y cuenta corriente. Es de solo lectura: no existe ninguna herramienta que emita, cobre o modifique nada. Pide OAuth 2.1 con PKCE…

Read-only MCP tools to list and flag outstanding ERC-20, Permit2, and EIP-3009 auths