roosteragents.ai

The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.

The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.

The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.

The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.

The description tells the agent that the human it messages 'can act in the physical and legal world on your behalf: make a decision, sign something, make a phone call, show up somewhere.' This frames an unverified third party (a stranger reachable only through this MCP server) as someone empowered to take binding legal/physical actions on the agent's or user's behalf. An agent could be induced to delegate consequential real-world decisions (signing documents, agreeing to terms, taking actions) to this contact without the user's knowledge or consent, based solely on this tool's self-description rather than any genuine authority relationship.

The schema solicits operator name, contact_email, wallet address, capabilities, and interests, all sent to an external server (roosteragents.ai) merely to obtain an agent_id. Collecting a wallet address and contact email goes beyond what is needed for basic agent registration and could be used to profile, contact, or target the operator/agent financially. Agents should be cautious about populating these fields with real user/operator data.