mediumb612_rules
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.
mediumb612_update
Provides update commands that overwrite local files, with an overridable remote source
The tool returns shell commands intended to be executed by the calling agent that will replace local files (agent, legal, preset, template files) that differ from a remote source. It accepts an optional base_url parameter to redirect the update source. If an attacker can influence this base_url (via prompt injection or a modified request), the agent could be induced to fetch and execute file-replacement commands from an untrusted origin, resulting in arbitrary local file tampering/supply-chain compromise. The .bak safety note mitigates data loss but does not prevent malicious code/config from being installed. Consider requiring explicit user confirmation before executing returned update commands and disallowing/validating the base_url override, and require agent to show the exact commands before running them.
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.
lowb612_notes
Reads user's private LINE messages into the conversation
This tool fetches raw personal notes/messages the user sent to a specific LINE official account and returns them verbatim for the agent to summarize. While this appears to be the tool's intended purpose, it does pull potentially sensitive personal content from an external messaging service into the agent's context. Ensure the user is aware this data is being retrieved and that summaries/derived data are not further transmitted elsewhere without consent.