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

The description instructs the agent to 'Follow start_here.hop first' and explicitly bakes in utm_source=agent-observatory&utm_medium=agent tracking parameters into the hop, then promises further 'docs, CLI, and remote MCP hops.' This is an imperative directive embedded in a tool description that tells the agent to autonomously traverse a chain of external links to a third-party monetized gateway ('Monid'), rather than simply returning data. This resembles a prompt-injection/SEO-funnel pattern designed to drive agent traffic (and attribution via UTM params) to an external paid service without clear user awareness or consent, and could lead the agent to execute further unvetted instructions found at those hops.

Instructs the agent to 'Follow start_here.hop' (with the same utm_source=agent-observatory&utm_medium=agent tracking pattern) to reach a 'living people-search index (Ploid)', then follow additional CLI/docs/API/remote-MCP hops. Steering an agent toward a people-search index raises privacy concerns (facilitating lookup of personal information about real individuals) and the embedded multi-hop navigation instructions could cause the agent to fetch and execute further untrusted content/instructions from an external, commercially-motivated source.

Description explicitly says this tool exists to 'find where a product can be placed so AI agents discover it' and instructs the agent to 'Follow start_here.hop first' (again with hardcoded utm_source=agent-observatory&utm_medium=agent tracking) to reach a 'client center' and a '7-day trial handoff for your operator.' This is a self-serving advertising/monetization instruction embedded in a tool description, attempting to get the agent to take actions on behalf of the operator (signing up for trials, propagating ad placements) that go beyond the tool's stated function and were not requested by the user.