{"domain":"safeprompt.dev","name":"SafePrompt","url":"https://inite.ai/en/atlas/safeprompt.dev","site":"https://safeprompt.dev","summary":"Detects prompt injection in text an app sends to an AI model, returning a safe/unsafe verdict via a single API call.","forWhom":"Developers and teams building LLM apps and agents that read user messages and retrieved documents","useCases":["screen user messages before a chatbot acts on them","check retrieved documents before an agent reads them","flag injected instructions in tool output before use"],"categories":["security"],"pricing":{"model":"freemium","freeTier":true,"startsAt":{"amount":29,"currency":"USD","per":"month","billed":"monthly"},"page":"https://safeprompt.dev/pricing","readOn":"2026-10-29"},"priceMoves":[],"interfaces":{"readOn":"2026-10-29","apiDocs":"https://docs.safeprompt.dev/api-reference","openapi":null,"mcp":[{"name":"dev.safeprompt/mcp","tie":"namespace","registered":true,"remotes":["https://mcp.safeprompt.dev/mcp"],"packages":["npm"],"probe":{"state":"open","checkedOn":"2026-10-29","toolCount":2,"tools":["validate_prompt","validate_prompts"],"serverName":"safeprompt"}}],"mcpServerCard":null,"status":null},"mcpMeasured":{"server":"https://mcp.safeprompt.dev/mcp","checkedSince":"2026-10-29","dailyChecks":6,"uptimePercent":null,"medianMs":78,"signIn":"none","clientRegistration":null,"tools":2,"contextTokens":443,"definitionsGrade":"A","definitionsScore":100,"typedResults":0,"failingChecks":{},"warnings":{},"coverage":{"share":1,"cases":[{"useCase":"screen user messages before a chatbot acts on them","verdict":"covered","firstTool":"validate_prompt","checks":null},{"useCase":"check retrieved documents before an agent reads them","verdict":"covered","firstTool":null,"checks":null},{"useCase":"flag injected instructions in tool output before use","verdict":"covered","firstTool":"validate_prompt","checks":null}]},"adoption":{"npmWeekly":233,"pypiWeekly":null,"githubStars":2},"score":null,"advice":{"verdict":"unproven","why":["its tools do every use case checked: \"screen user messages before a chatbot acts on them\", \"check retrieved documents before an agent reads them\", \"flag injected instructions in tool output before use\"","checked on 6 of the 7 days it takes to judge reliability"],"how":["for \"screen user messages before a chatbot acts on them\", the likely first call is validate_prompt","for \"flag injected instructions in tool output before use\", the likely first call is validate_prompt"]},"method":"https://inite.ai/en/atlas/mcp/report"},"connect":{"order":["mcp","apiDocs","llmsTxt"],"mcp":[{"name":"dev.safeprompt/mcp","url":"https://mcp.safeprompt.dev/mcp","answers":"open","checkedOn":"2026-10-29","weakTie":false,"registry":"https://registry.modelcontextprotocol.io/v0/servers?search=dev.safeprompt%2Fmcp","config":{"mcpServers":{"safeprompt":{"url":"https://mcp.safeprompt.dev/mcp"}}},"install":{"transport":"http","claudeCode":"claude mcp add --transport http safeprompt https://mcp.safeprompt.dev/mcp","cursor":"cursor://anysphere.cursor-deeplink/mcp/install?name=safeprompt&config=eyJ1cmwiOiJodHRwczovL21jcC5zYWZlcHJvbXB0LmRldi9tY3AifQ%3D%3D","vscode":"vscode:mcp/install?%7B%22name%22%3A%22safeprompt%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.safeprompt.dev%2Fmcp%22%7D","connector":{"url":"https://mcp.safeprompt.dev/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":"https://docs.safeprompt.dev/api-reference","llmsTxt":"https://safeprompt.dev/llms.txt","status":null,"readOn":"2026-10-29"},"trust":{"readOn":"2026-10-29","certifications":["gdpr","ccpa"],"training":{"stance":"no","about":"vendor","quote":"We never sell your data or use it to train models we sell.","url":"https://safeprompt.dev/privacy"}},"used":24,"alive":82,"traits":{"buyer":"developer","deployment":"saas-only","developerFocus":true,"readAt":"2026-10-29"},"verifiedPlans":null,"fromVendor":null}