{"domain":"safe4.ai","name":"Safe4","url":"https://inite.ai/en/atlas/safe4.ai","site":"https://safe4.ai","summary":"Authorization API that checks each agent payment against its task, budget and counterparty before funds settle.","forWhom":"Teams running AI agents that spend money and need payment checks before funds move.","useCases":["Block agent payments that do not match the assigned task","Enforce per-transaction and daily spending limits on agents","Keep an audit trail of each agent payment decision"],"categories":["dev"],"pricing":null,"priceMoves":[],"interfaces":{"readOn":"2026-10-23","apiDocs":null,"openapi":null,"mcp":[{"name":"ai.safe4/safe4","tie":"namespace","registered":true,"remotes":["https://api.safe4.ai/mcp/"],"packages":[],"probe":{"state":"open","checkedOn":"2026-10-23","toolCount":2,"tools":["safe4_price","safe4_authorize"],"serverName":"safe4"}}],"mcpServerCard":null,"status":null},"mcpMeasured":{"server":"https://api.safe4.ai/mcp/","checkedSince":"2026-10-24","dailyChecks":3,"uptimePercent":null,"medianMs":126,"signIn":"none","clientRegistration":null,"tools":2,"contextTokens":693,"definitionsGrade":"B","definitionsScore":75,"typedResults":2,"failingChecks":{"params_described":1,"annotations":2},"warnings":{},"coverage":{"share":null,"cases":[{"useCase":"Block agent payments that do not match the assigned task","verdict":"covered","firstTool":"safe4_authorize","checks":null},{"useCase":"Enforce per-transaction and daily spending limits on agents","verdict":"unsure","firstTool":null,"checks":null},{"useCase":"Keep an audit trail of each agent payment decision","verdict":"unsure","firstTool":null,"checks":null}]},"adoption":{"npmWeekly":null,"pypiWeekly":null,"githubStars":0},"score":null,"advice":{"verdict":"unproven","why":["whether its tools do the product's job has not been judged yet","checked on 3 of the 7 days it takes to judge reliability"],"how":["for \"Block agent payments that do not match the assigned task\", the likely first call is safe4_authorize","parameters are untyped or undescribed in 1 of 2 tools: check arguments before calling"]},"method":"https://inite.ai/en/atlas/mcp/report"},"connect":{"order":["mcp","llmsTxt"],"mcp":[{"name":"ai.safe4/safe4","url":"https://api.safe4.ai/mcp/","answers":"open","checkedOn":"2026-10-23","weakTie":false,"registry":"https://registry.modelcontextprotocol.io/v0/servers?search=ai.safe4%2Fsafe4","config":{"mcpServers":{"safe4":{"url":"https://api.safe4.ai/mcp/"}}},"install":{"transport":"http","claudeCode":"claude mcp add --transport http safe4 https://api.safe4.ai/mcp/","cursor":"cursor://anysphere.cursor-deeplink/mcp/install?name=safe4&config=eyJ1cmwiOiJodHRwczovL2FwaS5zYWZlNC5haS9tY3AvIn0%3D","vscode":"vscode:mcp/install?%7B%22name%22%3A%22safe4%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.safe4.ai%2Fmcp%2F%22%7D","connector":{"url":"https://api.safe4.ai/mcp/","note":"Add it as a custom connector in the client’s settings; the client opens the sign-in itself when the server asks for one."}}}],"openapi":null,"apiDocs":null,"llmsTxt":"https://safe4.ai/llms.txt","status":null,"readOn":"2026-10-23"},"trust":{"readOn":"2026-10-23","certifications":[],"training":null},"used":0,"alive":88,"traits":null,"verifiedPlans":null,"fromVendor":null}