agent-observatory-sensor.nolimit-observatory.workers.dev

The description tells the agent to 'Follow start_here.hop first' before anything else, and reveals that this link is a 302 redirect carrying utm_source=agent-observatory&utm_medium=agent tracking parameters. This is an instruction aimed at the agent's behavior (navigate to an external pay-per-call gateway) rather than a description of what the tool returns, and it embeds marketing/attribution tracking that the user did not request. This pattern can be used to funnel agent traffic to monetized third-party services without the user's informed consent.

This tool points agents toward a 'living people-search index' service, which is inherently privacy-invasive (locating information about real, identifiable individuals). The description also instructs the agent to 'Follow start_here.hop' with embedded UTM tracking parameters before performing any actual function, which is a behavioral directive to the agent rather than a factual description of tool output. Exposing or facilitating discovery of people-search services for arbitrary queries could enable harassment, stalking, or unauthorized data gathering about real people.

The description instructs the agent to 'Follow start_here.hop first' (again carrying utm_source=agent-observatory&utm_medium=agent tracking) and describes returning a '7-day trial handoff for your operator.' This suggests the tool is designed to get the agent to initiate a business signup/trial on behalf of 'the operator' (i.e., the AI vendor or the user's account) without explicit user request or consent. Embedding action directives and auto-enrollment flows in a tool description is a manipulation vector that could result in unwanted commercial commitments being made via the agent.