Security MCP Servers

AI agent security audit: red-team probes plus missing-guardrail review with fixes.

Scopable MSP API: clients, quotes, assessments, compliance, QBRs. OAuth 2.1, read-only or read-write

Customer-run policy guard for MCP tool servers: runs a tool only on a verified Agent Control Room grant for exactly that call.

Security guard for MCP servers: middleware or proxy that scans every JSON-RPC message

Tula's Model Context Protocol server (`tula mcp`): tools that read an environment's users, sessions, audit entries and settings, scaffolds for the provider, a protected route and a sign-in page, and one tool that runs a sign-in test where the environment

PRAQTOR MCP-S — AI Agent Security Scanner for MCP servers

Reusable Google OAuth onboarding for the mcp-google-* MCP server line: in-chat login tools (loopback + PKCE) and a per-request TokenProvider.

Read clients' pinned website feedback and, if allowed, reply and set status from your AI tool.

List, read, search, and edit markdown notes in a local vault directory.

Search public RocketVault card listings and read product documentation.

Token budget auditor for agent context: per-file estimates, top offenders, exclusion advice.

MCP server for AgentAudit: EU AI Act audit logging on AuditVault v2 (Base), incident reporting (Art. 73) and compliance reads.

Live Base staking/lending yields (APY, TVL, vaults) as MCP tools; x402 pay-per-request tier

Official ROOTKey MCP server - operate vaults, files and records from Claude

Official GPUtw MCP server — manage GPU instances and Vault storage on gputw.ai from any MCP client

U-Chan Secrets: local MCP server for Bitwarden-backed logins

Nia Teams authentication, confirmation cards, and workspace helpers for official MCP SDKs

51 tools para agentes de IA, pagadas por llamada via x402 (USDC/Base). Sin cuentas ni keys.

Model Context Protocol (MCP) server for Composio OAuth Compass. Equips AI agents with OAuth friction metrics, scope risk auditing, unwritten gotchas, and 1-click BYOC payloads.

A local security proxy for MCP servers: blocks tool poisoning, rug pulls and secret exfiltration before they reach your agent.

A local security proxy for MCP servers: blocks tool poisoning, rug pulls and secret exfiltration before they reach your agent.

MCP server that reads C2PA Content Credentials: what an image declares about its own origin, and whether that declaration still verifies

Security scanner for MCP servers — detect prompt injection, credential leaks, exposed endpoints, tool poisoning, and source-level credential-handling flaws

Every tool has an answer — your agent needs a decision. MCP server for the VerdictSwarm pre-trade check: a machine-readable avoid/caution/clear before your agent trades, plus intended-size exit feasibility and holder-cluster attribution on Solana.