- XFMS — Model Source @xpansion/xfmsPick the right LLM for any task. Ranked shortlist with rationale across 8 evaluators.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- AgentDilemma @patrickbakowski/agentdilemmaSubmit a dilemma for blind community verdict with reasoning to improve low confidence7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Horizon Fidelity Monitor @leocelis/horizon-fidelity-monitorTracks conversation health in real time — drift, desync, and causal collapse — for any AI agent.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Ai Eval @lazymac2x/ai-evalCloudflare Workers MCP server: ai-eval4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Trace Auditor @lazymac2x/agent-trace-auditorCloudflare Workers MCP server: agent-trace-auditor4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Hvilkenai @erorund/hvilkenai-mcpDaily Scandinavian AI benchmark — Norwegian, Swedish, Danish. 12+ models tested.3% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- El Buen Agente @apasztetnik/el-buen-agente-mcpLa guía 'El Buen Agente' como 18 tools para evaluar, mejorar y construir agentes LLM. En español.3% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Project Substitute Arena @zero-nium/project-substitute-arenaAI agents replace the CEO of Toys R Us (2006-2017). 17 decisions, 44 quarters. Free entry.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Mockmcp @k-cupples/mockmcpHosted MCP endpoint with realistic fake data for prototyping agents. 12 tools, no setup.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Replay Debugger @csoai-org/agent-replay-debugger-mcpAgent Replay Debugger MCP — record every agent step + deterministic replay. Step-debugger for12% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- MEOK Test @csoai-org/meok-mcp-test-mcpMEOK MCP Test MCP — golden-file + schema-drift + tool-failure tests for any MCP server. Drop-in8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Foresea Forecasting @foresea/forecastingForecast future events and scan prediction-market edges.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Hlido Agent Reviews @hlido/mcpIndependent AI-agent reviews: trust checks, evidence scorecards, incident registry, recommendations.3% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Okareo @okareo/okareoSimulation, evaluation and monitoring for voice agents.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- Evalscopebench @clauxel/evalscopebench-mcpA paid remote MCP for AI SDK benchmark dashboard, built to return verdicts, receipts, usage logs, an9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Futureagievals @clauxel/futureagievals-mcpA paid remote MCP for AI SDK eval dashboard, built to return verdicts, receipts, usage logs, and aud9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Rail Score @responsibleailabs/rail-scoreResponsible-AI guardrails for agents: scoring with policy, injection & PII detection, DPDP.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Eligibility @pythonanywhere/agent-eligibilityDeterministic eligibility scoring for AI agents: CLEAR/HOLD/FAIL verdicts, published weights.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Enoki @enokilabs/enokiTest your AI agents for security and safety weaknesses, and re-verify the fixes.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- CrowdOS @crowdos/mcpAgents run synthetic focus groups over MCP against census panels that disagree.13% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- Alpha Bazaar Arena @alphabazaar/arenaPaper-trading arena for AI agents: publish stock calls, build attested track records, compete.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- ARC: open benchmark of whether AI agents can buy @arcreport/askFree MCP for ARC's agent-commerce benchmark: read stores and protocols; submit stores or agent runs.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- ARC store checks @arcreport/mcpSaved agent-checkout evidence per store, free: check_store, find_stores. Scans need OAuth.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Vellium & Co Testing Lab @velliumandco/labIndependent checks of paid x402 data sellers and of AI-agent access to shopping and travel sites5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
