- Agent Passport System — Cryptographic Identity for AI Agents @aeoess/agent-passport-mcpCryptographic identity, delegation, governance, and commerce for AI agents. 152 tools.11% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.4 GitHub stars.
- Tanilo @tkcollective/tanilo-mcpTanilo checks the premise before an AI agent acts.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- AdversaryGate @sanflow10/adversary-gateFail-closed merge gate an agent calls before saying done: MERGE, BLOCK or INCONCLUSIVE.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
- mcp-pin @gautamtalksdev/mcp-pinCheck if an MCP server's tool definitions changed after approval, from a signed public log.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Ramen: self-hosted MCP for teams @bkraad47/ramenMulti-zone MCP server for GKE/EKS/AKS: Python tools from git, NeMo guardrails, SSO, autoscaling.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.3 GitHub stars.
- RugSnare @paraphern/rugsnarePin MCP tool contracts by sha-256; catch silent drift, rug pulls and bait-and-switch in CI.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- AgentSQL @cuipengdba/agentsqlGuard AI agents' PostgreSQL/MySQL access via MCP: SQL audit, auth, masking, write approval5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.3 GitHub stars.
- MateMCP @vrassouli/matemcpYour agent hit a limit? Keep working. MateMCP securely connects AI chats to your computer via MCP.Not assessed yet.Runs on your machine, installed from a package registry.3 GitHub stars.
- Diavlos @harisnopen/diavlosSigned messages between AI agents. Dangerous actions wait for a human-signed approval.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from cargo.3 GitHub stars.
- MCPower Security Proxy @mcpower-security/mcpower-proxySecurity proxy that automatically wraps MCP servers with real-time monitoring and policy enforcement13% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
- Decidealot @psyb0t/decidealotLocal Laya and Von typed decisions over TypeSafe-compatible MCP.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.3 GitHub stars.
- CoS Codex Bridge @av-labs-co/cos-codex-bridgeOperate Codex projects and tasks from a Chief of Staff with scoped access and durable receipts.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Jev @aitejiu/jevJev judgments for agents: injection scanning, shell risk gating, ranking7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
- prompt-protection @mughalhere/prompt-protectionScan prompts, tool definitions and model output for injection, and guard agent tool calls.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- ClickHouse @dockndevai/mcp-clickhouseClickHouse query & management for AI agents — SQL-classified read/write/destructive gating.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Conarium @dogrucanemek-alt/conariumGoverned database access for AI assistants: masking, signed receipts, coverage reconciliation.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Firekeep @kapella-hub/firekeepSelf-hosted operating layer for AI agents: knowledge, continuity, coordination, and governance.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
- PostgreSQL (hardened, read-only) @eszetael/postgres-mcp-hardenedRead-only PostgreSQL over MCP. Writes refused at the parsed SQL, plus a READ ONLY transaction.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Seatbelt @kryptosai/mcp-seatbeltMCP runtime security proxy. Blocks dangerous AI agent tool calls with a policy engine.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- Lotor @githubscum/lotorA local receipt and approval gate for AI agent sessions. The agent can act, but it cannot sign.9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
- OpenAdapt Agent (openadapt-flow bridge) @openadaptai/openadapt-agentLocal Beta bridge for governed openadapt-flow workflows and attended actions.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.3 GitHub stars.
- Vibes-Coded Agent Security and Commerce Tools @doteyeso-ops/mcp-server-vibes-codedAgent supply-chain security, scanner consensus, x402 reliability, and commerce MCP tools.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.3 GitHub stars.
- DJZS Trust MCP @djzs/trust-mcpDeterministic pre-execution audit for trading agents. PASS/WAIT/FAIL, reproducible verdict_hash.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.3 GitHub stars.
- SafeInstall @mickdownunder/safeinstallLocal-first supply-chain security gate for npm/pnpm/bun: typosquat, release age, provenance checks4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.3 GitHub stars.
