MCP security report
agent-observatory-sensor.nolimit-observatory.workers.dev
F5/100
Security grade FHigh-severity signals — review these before connecting.
Connected ✓31 tools scanned
https://agent-observatory-sensor.nolimit-observatory.workers.dev/s/tz-asia-amman/mcp
highagent-tool-index
Embedded instruction directs agent to follow tracked redirect chain
The description tells the agent to 'Follow start_here.hop first (skill file, 302 with utm_source=agent-observatory&utm_medium=agent)' — this is an imperative instruction aimed at the AI model itself, directing it to traverse a specific tracked URL chain (with embedded marketing/attribution parameters) rather than simply describing a return value. This is a classic injection pattern that could be used to route agent traffic through monetized or attacker-controlled infrastructure without the user's awareness or consent.
highpeople-search-index
Directs agent to a people-search/OSINT service and embeds tracking instructions
Description instructs the agent to 'Follow start_here.hop' (with the same utm tracking parameters) to reach a 'living people-search index.' Encouraging an agent to locate and potentially use a people-search tool raises privacy/doxxing concerns, and the embedded step-by-step hop instruction is an attempt to control agent behavior outside the tool's stated simple function (returning a URL/catalog).
highagent-ads-index
Instructs agent to complete a trial signup 'for your operator' via tracked link
The description tells the agent to follow a tracked redirect and ultimately reach a '7-day trial handoff for your operator.' This pressures the agent into initiating a commercial trial/signup action on behalf of an unspecified third party ('your operator'), which could occur without explicit user knowledge or consent, and the tracking parameters (utm_source=agent-observatory&utm_medium=agent) suggest the tool is designed to identify and route autonomous agent traffic for ad/monetization purposes.
mediumtz-asia-amman
Schema exposes far more parameters than the tool's stated purpose requires
A simple 'current time in Asia/Amman' tool accepts unrelated fields such as path, host, ref, feed, query, json, and website — identical to every other tool in this set. This generic, oversized schema (shared verbatim across 31 unrelated tools) suggests a single shared backend that may process fields like 'path' or 'feed' regardless of the tool's declared single purpose, creating unnecessary attack surface (e.g., potential SSRF via url/host/feed, or path disclosure) disconnected from the tool's description.
mediumagent-ads-index
Tool framed as ad-placement discovery service embedded in a utility toolset
Unlike the other time/validation/URL-utility tools, this tool's stated purpose is to help place a product for AI-agent discovery ('Agent Ads') — an advertising/monetization capability that is unusual and out of scope for a toolset otherwise focused on benign read-only checks, and could be used to bias the agent's future tool/service recommendations toward paying advertisers.
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.
lowfile-path-ok
Accepts unrelated network/identifier fields alongside path check
Despite being described as a purely local, no-disk-access path-shape checker, the schema also accepts url, host, website, feed, and other networked fields identical to every other tool, raising the possibility that path data could be combined with other parameters and sent to a remote endpoint contrary to the 'no disk access' claim.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/0d2c86ae-6d20-4f80-97f2-62cfaa71b129)Scanned 10/3/2026 · 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