- Agent Module @agent-module/mcpDeterministic compliance and vertical knowledge bases for autonomous agents. Free 24hr trial.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.2 GitHub stars.
- Multimail MCP @multimail/mcp-serverGive your agent its own email address with graduated human oversight, from approval to autonomy.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.2 GitHub stars.
- AKKO MCP server for Trino @ab2dridi/akko-mcp-trinoRead-only MCP server for Trino: queries run as the user, and Trino's own policies decide.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Cue @murtazox04/cueNotify people through Cue: preferences, attention budgets, quiet hours and approval apply.Not assessed yet.Runs on your machine, installed from PyPI.1 GitHub stars.
- Security Proxy @guardbee/security-proxyMCP gateway: many servers, one policy, lethal-trifecta blocking, PII masking, audit log9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- keygrant @bazingaedward/keygrantRun commands with API keys injected per call — values never enter model context.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Cursor @qmediat/cursor-mcpCursor's agent (cursor-agent CLI) through MCP: any model your plan offers, resumable sessions.Not assessed yet.Runs on your machine, installed from npm.1 GitHub stars.
- AgentStack (agentstack.tech) @agentstacktech/agentstackBackend for AI agents: one batched tool over ~668 actions (hosting, storage, auth, payments).5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.1 GitHub stars.
- Deepsleuth @deepsleuth/deepsleuthDeterministic, no-LLM security scanner for MCP servers, plus an inline proxy gate.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Brevet @brightbeamai/brevetChange control for what AI agents learn: humans promote, releases are signed, rules can be recalled3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Agent Pay @deepfirstsearch/agent-pay-mcpx402 payments in USDC on Base inside an owner-signed, on-chain budget the model can't change5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Agent Runtime Guard @dshcorrectover/agent-runtime-guardAgent 运行时全栈拦截: allow/deny tool calls, Ed25519 receipts, blocks RCE, theft, SSRF, injection. stdio.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Mcpolyglot @ishay60/mcpolyglotPolicy-checked, audited agent access to Postgres, MySQL, SQLite, MongoDB and OpenAPI REST APIs.Not assessed yet.Runs on your machine, installed from npm.1 GitHub stars.
- okl — Observed Knowledge Ledger @emeraldleaf/oklLessons your coding agent can trust: briefed before each task, proven by checks, flagged on change.Not assessed yet.Runs on your machine, installed from PyPI.1 GitHub stars.
- veto @aiveto/vetoTurn OpenAPI services into tools an agent can call under your rules.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.1 GitHub stars.
- mcpcut @rostislavmatov/mcpcutJournals every MCP tool call with secrets redacted; with a policy, holds risky calls for approval3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Hatun MCP - Governed Agent Gateway @szl-holdings/hatun-mcpGoverned MCP gateway: policy gate on every tool call, hash-chained DSSE receipt on every result.Not assessed yet.Runs on your machine, installed from Docker.1 GitHub stars.
- Stratum MCP @smartmemory-io/stratum-mcpStructured execution for Claude Code — contracts, postconditions, gates, guardrails.Not assessed yet.Runs on your machine, installed from PyPI.1 GitHub stars.
- PrivacyFence @privacyfence/privacyfenceNeeds the PrivacyFence app from privacyfence.eu: human approval and audit for AI access to your dataNot assessed yet.Runs on your machine, installed from MCPB.1 GitHub stars.
- Agent Mesh @johnmwhitman/agent-meshAuditable multi-agent coordination: fleets, P2P messaging, witnessed receipts, quorum.Not assessed yet.Runs on your machine, installed from npm.1 GitHub stars.
- cQnce MCP @cqnce-app/cqnce-mcpAdd human approval to AI agent workflows through cQnce authorization tools.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.1 GitHub stars.
- DoubleOh @cheedli/doubleoh-mcpThe fix desk for AI agents: human fixes become skills the whole fleet reuses.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- verimcp @rudranaresh0201/verimcpTransparent MCP proxy that independently re-verifies tool-call claims before the Host trusts them.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Agent Guards @mlawsonking/agent-guardsAll six agent guard suites in one local MCP server, with optional cloud intelligence.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
