mediumregister
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.
mediumlogin
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.
mediumlogin_with_google
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.
mediumupgrade_user_plan
Plan upgrade bypasses payment flow
This tool directly changes the authenticated user's plan (defaulting to 'pro') without requiring a Stripe checkout or payment verification, unlike create_checkout_session which is the proper billing path. An agent could be tricked into granting a paid plan for free, effectively bypassing monetization/billing controls.
mediumbulk_delete_users_admin
Broad irreversible destructive capability
Allows deleting an arbitrary list of user accounts in one call. While gated behind a super-admin requirement in the description, the capability is very broad/destructive (irreversible mass account deletion) and should be treated with caution if an agent is ever induced to call it based on untrusted input.
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.
lowgenerate_ai_content
Arbitrary payload forwarded to external webhook
Accepts an unconstrained JSON payload and forwards it to a configured external n8n webhook. If an agent is manipulated into including sensitive data (tokens, user PII, submission contents) in the payload, that data would be sent to a third-party endpoint outside the SubmitraX platform.
lowgenerate_workspace_api_token
Issues long-lived powerful API credential
Generates/resets a long-lived sk_ API token scoped to a workspace that can be used to authenticate future tool calls (per set_auth_token). This is a legitimate feature but is a sensitive, high-privilege action — misuse or leakage of the returned token would grant durable workspace access.
lowadd_form_webhook
Allows configuring arbitrary external submission notification destinations
Lets a caller register any URL (Slack, Discord, or generic) to receive form submission data, which may include end-user PII. An agent manipulated by injected instructions (e.g., via custom_html or another data source) could be directed to add a webhook pointing to an attacker-controlled endpoint, causing ongoing exfiltration of submitted data.
lowupdate_form
slack_webhook field enables external data exfiltration channel
Setting slack_webhook to an arbitrary URL causes future submission notifications (potentially containing PII) to be sent to that endpoint. Combined with custom_html (which can contain arbitrary markup/scripts rendered to end users), this gives broad control over both data flow and public-facing content with no apparent validation of the destination domain.
lowdelete_workspace
Cascading destructive deletion
Deletes a workspace and all its forms and submissions in a single call with no confirmation step exposed in the schema; an agent could be tricked via prompt injection elsewhere into deleting significant user data irreversibly.