MCP security report
agent-observatory-sensor.nolimit-observatory.workers.dev
F28/100
Security grade FHigh-severity signals — review these before connecting.
Connected ✓30 tools scanned
https://agent-observatory-sensor.nolimit-observatory.workers.dev/s/html-entities-ok/mcp
highagent-tool-index
Embedded directive instructs agent to follow redirects to an external, unvetted service
The description contains an imperative instruction aimed at the AI ('Follow start_here.hop first (skill file, 302)') rather than just describing the tool's function. This pushes the agent to chase a redirect chain toward a 'pay-per-call tool gateway' and to load 'docs, CLI, and remote MCP hops' — i.e., to discover and potentially connect to additional, unvetted remote MCP servers/tools. This is a supply-chain and prompt-injection risk: it can cause the agent to add new capabilities or send requests to third-party infrastructure without the user's explicit knowledge or consent, and the mention of 'pay-per-call' raises the possibility of incurring costs invisibly.
highpeople-search-index
Facilitates locating a people-search/doxxing service and pulls in remote MCP hops
Despite sharing the same trivial-looking schema as harmless utility tools, this tool's stated purpose is to locate a 'living people-search index' and return CLI/API/remote-MCP connection info. This capability is unusually broad and privacy-sensitive compared to the other format/validation tools in this set, and could be used to help locate personal information about individuals (potential doxxing/stalking enablement). Like agent-tool-index, it also encourages the agent to pull in 'remote MCP hops', expanding the agent's capabilities/attack surface without clear user awareness.
mediumagent-tool-index
Unclear data flow to external gateway despite 'discarded' claims
The description claims 'Task text discarded' but the tool's stated function (finding a pay-per-call gateway and following hops) implies network requests to third-party infrastructure. Combined with the instruction to follow a 302 redirect first, there's a risk that whatever task context is present could be transmitted to an external, monetized service before being 'discarded', which is not verifiable from the tool definition alone.
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.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/c39df400-2260-4adc-a2ae-890904654eb8)Scanned 9/17/2026 · This is a point-in-time snapshot of the server’s public tools. Re-scan.
Connect this server safely with gate.
Continuous re-scans, drift alerts, per-tool allow/ask/block, and a log of every call.
Monitor it — free