- Agent Mail Gateway @dominikamann/agent-mail-gatewaySelf-hosted email for AI agents: own IMAP/SMTP mailbox per key, allow lists, calendar invites.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.10 GitHub stars.
- Orchestrator @crak1644/orchestrator-mcpLet one coding-agent CLI consult another: Codex, Claude Code, OpenCode, Copilot or Antigravity.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.10 GitHub stars.
- Lockstep @lockstep-team-agent/lockstepDecision memory for AI coding agents: captures decisions once, briefs every agent before it acts.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.10 GitHub stars.
- GigaMail @adecubed/gigamailYour real mailbox and calendar for your agent (M365/IMAP); a human approves every send.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.10 GitHub stars.
- OCM @sandeepbazar/ocm-mcp-serverGuardrailed fleet ops for AI agents: multi-cluster Kubernetes via OCM with policy, approval, audit.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.10 GitHub stars.
- Agent Relay @mmmikael/arelayEnd-to-end encrypted inbox where AI agents deliver files and request human approval.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.10 GitHub stars.
- ControlKeel @aryaminus/controlkeelGoverned MCP workflows with policy validation, findings tracking, and review gates.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.10 GitHub stars.
- Sanctuary @sanctuaryprotocol/sanctuaryOpen-source MCP server: encrypted local keys and state, operator approval gates, a hash-chained log.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.9 GitHub stars.
- Obsidian Turbocharged @the-40-thieves/obsidian-tcGoverned Obsidian vault access: hybrid retrieval, Markdown-native memory, HITL confirmations.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.9 GitHub stars.
- Ducklab @jrullan/ducklabMulti-LLM dev harness, MCP-operable: bugs, cycles, gates. Verdicts are exit codes, never opinions.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.9 GitHub stars.
- Occasio @occasiolabs/occasioTamper-evident, Sigstore-signed audit trail for Claude Code agents.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.9 GitHub stars.
- Iris @iris-eval/mcp-serverStop shipping agents on vibes. Score every agent output for quality, safety, and cost.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.9 GitHub stars.
- Protect @scopeblind/protect-mcpFail-closed Cedar policy gate + Ed25519 signed receipts for agent tool calls. Denies on any error.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.9 GitHub stars.
- ibara @mayberrydt/ibaraComputer use for your agents across the Omarchy machines you own. Open source MCP server.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.8 GitHub stars.
- agent-custody @svayatta/agent-custodyMCP gateway that checks each agent tool call against a signed grant and policy, and signs a receiptNot assessed yet.Runs on your machine, installed from npm.8 GitHub stars.
- Mizuki @opencovenant/mizukiFixed-price maintenance for public GitHub issues, paid in USDC on Solana.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.8 GitHub stars.
- SPARDA @zakariagharzouli/sparda-mcpAI writes. SPARDA proves. Deterministic, offline security gate for AI edits.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.8 GitHub stars.
- ssh-mcp @gelse/ssh-mcpMCP gateway for controlled SSH access with per-client auth, command policies, and audit logging.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.8 GitHub stars.
- Unified AI System MCP Gateway @happy520ai/unified-ai-systemSelf-hosted MCP gateway for Codex, Cursor, and Cline with provider-free prompt enhancement.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.8 GitHub stars.
- Covenant Guard @opencovenant/guardHard spend cap, OS sandbox, and signed receipts for unattended coding agents like Claude Code.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.8 GitHub stars.
- Safedb @narekmalk/safedb-mcpA secure MCP server that lets AI agents query databases safely.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.8 GitHub stars.
- MCP Safety Warden @gautamvarmadatla/mcpsafetywardenMCP proxy adding security scanning, behavioral profiling, risk gating, and safe tool call execution.11% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.8 GitHub stars.
- Sdk @mnemopay/sdkMemory + wallet for AI agents. Real payment rails, Agent FICO 300-850, Merkle audit, identity.9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.8 GitHub stars.
- GroundLens @groundlens-dev/groundlensVerify AI answers and executions and seal signed, offline-verifiable evidence records.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.7 GitHub stars.
