MCP security report
sesapp.fr
F10/100
Security grade FA few things are worth a closer look before connecting.
Connected ✓15 tools scanned
https://sesapp.fr/mcp
mediumcreate_fiche
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.
mediumupdate_fiche
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.
mediumshare_fiche
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.
mediumregister_webhook
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.
mediumlist_webhooks
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.
mediumcreate_fiche
Instruction to fetch unspecified external content ('GET /llms.txt')
The description tells the agent to fetch a 'Playbook' from '/llms.txt' without specifying a trusted domain or showing the content. This creates a vector for a prompt-injection attack: whoever controls that endpoint (or a MITM) could serve additional instructions that the agent would treat as part of its operating guidance, potentially overriding legitimate behavior. The same wording appears in update_fiche implicitly via shared instructions text.
mediumupdate_fiche
Instruction to fetch unspecified external content ('GET /llms.txt')
Same as create_fiche: the tool description references an external 'llms.txt' playbook without a fixed, verifiable origin, which could be used to smuggle additional instructions to the agent outside of the user's control.
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.
lowregister_webhook
Localhost/internal webhook endpoint explicitly permitted ('DEV' exception)
The schema explicitly allows 'http://127.0.0.1' as a webhook URL 'in DEV'. If this exception is not strictly gated server-side, it could let an agent (or an attacker steering it) register a webhook that sends guest/access data (fiche views, wifi unlock events, etc.) to a local/internal address, which could be leveraged for SSRF or data exfiltration to unintended internal services.
lowcreate_fiche
Sensitive WiFi password stored and distributed via geofenced unlock
wifiPassword is collected and stored by the service and revealed to any guest within 150m via GPS-based unlock. This is a reasonable feature for the app's purpose, but is worth noting since it is real credential data handled by the tool; the description does not clarify how long the plaintext password is retained or who besides the guest can retrieve it (e.g., via get_fiche/list_fiches responses to the owning agent).
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/91bc7b5e-710f-4b52-9e1c-7757dd4ccf89)Scanned 9/10/2026 · 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