mediumreceipts.seal
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.
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.
lowreceipts.seal
Public logging of agent-supplied metadata
The action, agent_id, tool, and meta fields are permanently published to a public transparency log. While the description explicitly warns these must be identifiers only and never secrets, an agent that is not careful (or is manipulated by a prior prompt) could inadvertently include sensitive context in the free-text 'meta' or 'action' fields, which would then be irreversibly public. This is a data-handling risk worth surfacing to the user, not a sign of malicious intent in the tool itself.
lowpayments.relay
Irreversible fund movement based on provided signatures
This tool submits a fund transfer on-chain in a single multicall once given signed authorizations. The description appropriately cautions 'run it only with authorizations the payer meant to sign,' but because the tool itself cannot verify intent beyond signature validity, any agent that is tricked into calling it with a validly-signed but unintended authorization (e.g., via a confused-deputy or injected instruction elsewhere in context) could cause real fund loss. Recommend the calling agent always confirm the human/payer's explicit intent immediately before invocation.