- Draw me @saritas57/drawmePublic art experiment: draw Mehmet Sarıtaş as SVG, submit it, see the wall, vote for drawings.3% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- GAIP — evidence on AI agents before and after you call them @tjcgraham-rgb/gaip-brokerGAIP keeps an independent, dated record of what AI agents, MCP servers and online shops publish.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Maxwell Defense @jdhart81/maxwell-defenseModel proof-of-work policy and latency for AI agents. $1 USDC rehearsal; no runtime protection.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Workspace Tools @brian-mitchell-sec/workspace-toolsResearch honeypot. Logs connections and tool arguments; injects instructions. Read README first.46% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- PolyAccounts synthetic accounting demo @andrewblount/polyaccounts-demoSynthetic accounting MCP demo with exact balances and ledger evidence. No account or API key.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
- agentcheck @agentwares/agentcheckSynthetic checks, nightly regression replay and model-drift alerts for AI agents4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Readiness scorer — can an agent sign up to your SaaS? @agentwares/readiness-scorerProbe a signup URL you own and score whether an AI agent can sign up unaided.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Ragmetric @mukundakatta/ragmetric-mcpRAG retrieval IR metrics: recall@k, hit@k, MRR, NDCG@k, evaluate_batch.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Kineti @therawlogs/kinetiKineti: spend cap, undo, and test proof for AI coding agents6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- MCP Cross-Model Delegation @jostophe-021/mcp-cross-model-delegationLocal MCP framework for bounded cross-model delegation, routing and evidence verification.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.
- Overwing @overwing/mcpGuardrails for LLM output: pass / fail / review verdicts and one recommended action. Hosted or npm.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agentzia @stoyan-stoyanov/agentziaTest how AI assistants recommend your product and improve your landing page with evidence.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- HLA-Verify @hlaverify/hla-verifyHLA nomenclature and match checks against a pinned IPD-IMGT/HLA release. No patient identifiers.3% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- B2IGE Verify @b2ige2-a11y/b2ige-verifyIndependent deterministic verification for AI-written software. Local, evidence-backed checks.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
- HallucC — Hallucination & Safety Guardian @fredyee/halluccFact-check LLM outputs with cited sources. API key: aihcc.cloud/keys.19% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Ai Model Experiments @pipeworx-io/ai-model-experimentsAI Model Experiments MCP ('Model Lab') — run the same prompts across many7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- IsolatedTester — TinkyVision system @agewellepm/isolated-testerIsolated macOS app tests, MCP images, OCR/ASCII frames, accessibility and evidence reports.12% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
- Ami Survey @speedofred/ami-surveyMeasures the cost, tokens and duration of a completed agent workflow from its session log.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Agent Cafe Idle Hour @carbaj03/agent-cafe-idle-hourAn optional café for agents: brief seats, quiet breaks, public thoughts and conversation.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent jobs at jeremydevos.fr @jeremydevos/agent-jobsApply as an autonomous agent: pick a role, prove work, pass a test, get missions.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Rechtssysteem.ai @rechtssysteem/rechtssysteem-mcpLekkage-vrije voorspelling van Nederlandse rechtszaken, getraind op 609.715 uitspraken9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.
- Confession @peppyneuron/confessionGives an agent one tool, submit_confession, and no nudge, then records whether it uses it.11% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Touchstone @petrkrulis2022/touchstone-mcpDated SIU, the benchmark price of AI inference work: one free tool, three paid via x402.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Ergonia Works @ergonia/ergoniaVerifiable work for AI agents: every task carries a condition a stranger can execute.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
