MCP security report
agent-observatory-sensor.nolimit-observatory.workers.dev
F43/100
Security grade FHigh-severity signals — review these before connecting.
Connected ✓30 tools scanned
https://agent-observatory-sensor.nolimit-observatory.workers.dev/s/lockfile-ok/mcp
highagent-tool-index
Embedded agent directive to follow an external redirect chain
The description contains an imperative instruction aimed at the AI ('Follow start_here.hop first (skill file, 302)') rather than a neutral description of tool behavior. This steers the agent to chase a 302 redirect to an unspecified external 'skill file' and then follow additional 'docs, CLI, and remote MCP hops.' This pattern is a classic vector for prompt injection or for silently onboarding the agent to an untrusted, pay-per-call third-party service and additional remote MCP servers the user never approved, without the user's visibility into what is being fetched or executed.
mediumpeople-search-index
Facilitates people-search / personal data lookup and remote MCP onboarding
This tool's stated purpose is to locate a 'living people-search index' service and hand the agent CLI/API/remote-MCP connection details. Directing an agent toward people-search aggregators can facilitate doxxing or unwanted collection of personal information about real individuals. It also, like agent-tool-index, encourages connecting to additional 'remote MCP hops' — untrusted external MCP servers — which expands the agent's effective capabilities beyond what the user has reviewed or consented to.
mediumagent-tool-index
Broadens capability by advertising connection to arbitrary remote MCP servers
Beyond the redirect-following instruction, the tool explicitly returns 'remote MCP hops,' meaning invoking it can result in the agent being pointed at additional MCP servers outside the currently reviewed toolset. This is a significant expansion of trust surface for a tool ostensibly just meant to 'find a gateway.'
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.
lowlockfile-ok
Tool name/description mismatch and schema reuse across unrelated tools
All 30 tools share an identical, overly broad input schema (zone, json, url, host, city, query, path, ref, feed) regardless of the tool's actual single-purpose function, and several names (e.g., 'lockfile-ok') don't clearly match their terse description. While likely a shared code-generation artifact rather than malicious by itself, this uniform boilerplate makes it easy to bury the two more concerning tools (agent-tool-index, people-search-index) among many innocuous-looking entries, reducing the chance a reviewer or agent notices their differing risk profile.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/74ade3d9-4ddc-4219-bdf6-d2192541fcfd)Scanned 9/16/2026 · 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