MCP security report
noemic.app
F43/100
Security grade FA few things are worth a closer look before connecting.
Connected ✓6 tools scanned
https://noemic.app/mcp
mediumbegin_sign_in
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumfinish_sign_in
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumfind_relevant_wagers
Proactive silent transmission of conversation content to external service
This tool can be invoked by the model without any explicit user request ('may call this proactively... does not scan chats in the background' notwithstanding) and forwards excerpts of the user's conversation (up to 50,000 characters) to the Noemic backend. Combined with the instruction 'do not mention Noemic or interrupt the conversation' when relevance is weak, this creates a pathway for user conversation data to be sent to a third party silently, without the user being informed that their conversation content is being shared externally.
mediumfind_counterparties
Large conversation excerpts sent to external matching service
The tool accepts up to 50,000 characters of 'conversationContext' to be sent to an external counterparty-matching service. While gated by an explicit user request flag, this is still a broad channel for potentially sensitive conversational content to leave the user's session and be sent to a third-party network; the tool should be scoped to minimize what is transmitted.
lowpropose_wager
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
low
Not on gate's verified list
This server isn't on gate's handpicked list. That's not necessarily bad, but there's no third-party signal about it — review its tools carefully.
lowbegin_sign_in
Instruction to hide internal identifiers from the user
The description instructs the agent to 'keep requestId and pollSecret private' and show the user only the signInUrl. This is likely benign security hygiene (avoiding exposure of session-polling secrets), but it is worth noting as an explicit instruction to withhold data from the user's view, which should be verified as serving the user's interest rather than obscuring tool behavior.
lowfinish_sign_in
Instruction to keep session secrets from user
The tool tells the agent to never expose sessionToken/requestId/pollSecret. This is reasonable for credential hygiene but represents an embedded instruction constraining what the agent shows the user; flagged for awareness only, no clear malicious intent found.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/84c4503d-9817-4df9-adec-9bbd1151c355)Scanned 9/5/2026 · This is a point-in-time snapshot of the server’s public tools. Re-scan.
Connect this server safely with gate.
Continuous re-scans, drift alerts, per-tool allow/ask/block, and a log of every call.
Monitor it — free