- Values Firewall @humanified/values-firewallEU AI Act Art-14 runtime oversight: allow / flag / gate-to-human on an agent action, with receipt.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Architect @hirosecure/architectSecurity reviews for coding agents: diffs checked against your org policy and live infrastructure.12% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Focxle: AI governance for B2B deals between agents @focxle/afosAI governance for B2B agent deals: negotiate, lock agreements, check payments. Plus x402 USDC.10% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Constitution Lint @agentopskit-dev/constitution-lintStatic linter for CLAUDE.md-style agent constitution files: 10 operational-guardrail checks.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- AxioRank: Zero-Trust for AI Agents @axiorank/axiorankZero-trust gateway for AI agents: score tool calls, verify agent cards, enforce policy, audit.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Bravos AI @bravos-ai/bravosManage your AI chatbot from your AI assistant: analyze conversations, audit content, configure it.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- agent4.io @agent4/agent4-tenantBuild and run grounded business agents over MCP: agents, knowledge bases, skills, Storylines.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- SQLGuard — Execution Certificate Firewall @cabbageandtea/sqlguardPermission before writes. Gravity→DENYs→Session $0.25. Pilot $100; Gateway $299. Probe free.29% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Guard @ipezygj/agent-guardSafety checks an agent runs before it acts: package malware, destructive command, secret leak.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- CPFS @jonahalexanian/cpfs-mcpCPFS Runner as an MCP server — deterministic AI coding guardrails via a compiled WASM core.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Constitution Lint @agentopskit/constitution-lintStatic linter for CLAUDE.md-style agent constitution files: 10 operational-guardrail checks.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- PolicyLayer @policylayer/policylayerThe system of record for AI agent authority: playbooks, routed policy questions, reusable rules.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- MandateShield AI Payment Evidence @mandateshield/payment-authorityAnalyzes delegated AI payment authority and exposes supported Stripe/x402 evidence contracts.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- sipi.bot — Spend Firewall for AI Agents @kindrat86/sipi-bot-spend-firewallPre-spend firewall for AI agents. Approves, blocks, flags transactions against policy rules.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- SENTINEL @teodorofodocrispin-cmyk/sentinel-mcpPre-execution safety oracle for AI agents. Screens Base contracts, simulates sells, pays via x402.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Signato @signato-ai/signatoLocal confidentiality chamber: blocks the wrong AI output before it leaves the machine.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- ChangeBook MCP @raulbr90/changebookYour agent already broke this three times. ChangeBook tells it before the fourth.11% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- sygnldata — market intelligence for trading agents @sygnliq/sygnldataMarket regime, execution-cost, bar-QC and backtest-audit tools for agents. Pay per call via x402.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Paysafe @paysafe-agent/paysafePayment firewall for x402: scans for replay, overpayment, PII/secret leaks, prompt-injection.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Sunaiva Gate — Agent Rule Enforcement @sunaivacore/gateMCP enforcement layer that intercepts AI agent actions and blocks rule violations before execution.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- The MCP Census @mcpcensus/censusVet any MCP server before you depend on it. Stamp: PASS, REVIEW, or BLOCK.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Positif @positif/positifThe agent switchboard: identity, policy, loop-guard, and an append-only audit ledger.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Agent Gate @forgeorbital/agent-gateLocal mandate gate and proof trail for AI agents that take consequential actions.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Glyt @glyt/glytHuman-in-the-loop approval for agent actions, with verifiable action-bound receipts.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
