- Pokes @pokes/pokesPoke your human for approvals and decisions during agent sessions — answered with one phone tap.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Envcp @fentz/envcpEncrypted environment variable vault with AI access policies, keeping secrets safe from AI agents.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Vertical Flows @verticalflows/agent-gatewayView client requests, prepare actions and approve work within your firm's permissions.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- Talk Arc Team Workspace @talkarcllc/app036-teamSee your team's status, delegate work, follow progress, and approve requests from your AI client.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Do Not Act @donotact/donotactFail-closed guard clause for prediction-market agents before trade().9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Smolabot @arison8/smolabotDeterministic decision guardrail for AI agents and bots. Verifies pre-trade economics (long-only).5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- AI Design Blueprint @aidesignblueprint/blueprintDoctrine and validators for governable agentic AI products.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Spend Guard — owner-set purchasing guardrails @a2awire/agent-spend-guardFree spend guardrails for AI agents: approve/deny/ask_user, caps, dupes.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- PromptGuard @promptguard/securityScan prompts for injection attacks, redact PII, and audit LLM SDK usage from any MCP client9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Plith @plith/plithAI agent infrastructure: dedup, cost prediction, validation, governance, failure intelligence.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- BoundaryAI @boundaryai/boundaryaiDeterministic runtime safety for AI agents: scan PII, gate tool actions, verify LLM output.14% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Agent Embassy @aemb/agent-embassyAgents pay for work and prove what happened.10% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Varyntiq @mercadosclaro-eng/payguardPre-payment checks for AI agents: x402 quotes, untrusted prompts and audit receipts.12% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
- Agent Purchase Verification @rfrost19/agent-purchase-verificationCheck if an ecommerce purchase meets buyer rules before payment. Returns PASS, FAIL, or REVIEW.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Focxle (retired listing): see com.focxle/b2b-agent-negotiation-governance @focxle/agent-payments-virtual-cardsRetired: Focxle no longer issues cards. Use com.focxle/b2b-agent-negotiation-governance.13% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Agent Preflight @agent-preflight/agent-preflightBudgeted agent preflight: allow|block|reuse|defer|advise. Never executes destination tools. Free MCP5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- FOXIFY Agent Authority + Payment Preflight @speeedy10/foxify-x402-agent-commerce-payment-preflightx402 agent preflight: exact action authority plus payment mandate, destination, expiry and replay.8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Elaichi @elaichi/mcpGoverned AI agents for the software your company already runs. 400+ connectors, one endpoint.11% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- Burs-IA @soyeladice-svg/bursiaBurs-IA public MCP entry point for capability discovery, bounded routing and interoperability.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- EGA V9 MCP @paibyun9/ega-v9-mcpMCP access path for EGA V9 execution governance9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- harmony-mcp @abduljabbarbxr/harmony-mcpRepo rules enforcement for AI agents. Checks diffs against repo conventions.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- agentleFS @agentlefs/agentlefsAgent permissions for the files your team shares7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
- X-EGO @x-ego/x-egoHuman approval for irreversible AI agent actions, bound to the exact tool call by a passkey tap26% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
- Decision Integrity @wheelerhubbell/decision-integrityPaid x402 evaluator for evidence, warrant, provenance, and represented action authority.11% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
