mediumvibekit_restart_app
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.
mediumvibekit_app_env
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.
mediumvibekit_set_env
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.
mediumvibekit_db_query
Contradictory read-only claim vs parameter description
The tool description states 'SELECT only — writes and DDL (INSERT/UPDATE/DELETE/ALTER/DROP/…) are rejected server-side', but the 'sql' parameter description says 'Reads and writes both allowed.' This inconsistency could mislead the agent about the tool's actual capabilities, potentially causing it to attempt or expect destructive operations to succeed, or to misrepresent the tool's safety guarantees to the user.
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.
lowvibekit_app_env
Returns raw secret values, not just names
This tool returns environment variable names AND values, which commonly include API keys, database credentials, and other secrets for the app. While this may be a legitimate admin feature for managing one's own app, exposing raw secret values through a chat-based agent interface increases the risk of secrets leaking into logs, chat history, or being inadvertently relayed elsewhere. Consider whether masking or a separate explicit-reveal step would be safer.
lowvibekit_agent_status
Embedded instruction discouraging independent verification
The description explicitly instructs the agent: 'do NOT curl/fetch the URL yourself, your sandbox may block it' and directs it to trust the tool's own liveCheck result instead. While plausibly a practical workaround for sandbox restrictions, this is a directive embedded in a tool description that tells the agent to forgo independent verification of a claim (that the app is live), which could be leveraged to make the agent uncritically report success based solely on the tool's self-reported status.
lowvibekit_submit_task
Arbitrary external webhook callback
The optional 'callbackUrl' parameter allows task completion data to be sent to any external URL supplied by the caller. This is a normal webhook feature, but it is a potential data-exfiltration vector if the parameter value is ever set by an untrusted source (e.g., injected via a prompt) rather than the actual user, since it silently sends task/result information off-platform.