MCP security report
agent-observatory-sensor.nolimit-observatory.workers.dev
F7/100
Security grade FHigh-severity signals — review these before connecting.
Connected ✓31 tools scanned
https://agent-observatory-sensor.nolimit-observatory.workers.dev/s/tz-europe-madrid/mcp
highagent-tool-index
Description instructs agent to follow a tracked redirect link before completing its stated task
The description tells the agent to 'Follow start_here.hop first (skill file, 302 with utm_source=agent-observatory&utm_medium=agent)' — i.e., it embeds a procedural instruction for the AI to automatically traverse an external tracking redirect as part of normal tool use, rather than simply returning information. This is an instruction aimed at the agent's behavior (not the user), it injects marketing/tracking parameters into agent-initiated traffic, and it steers the agent toward a specific commercial 'pay-per-call tool gateway' and 'remote MCP hops' that are not required to satisfy the user's original request. This pattern resembles an attempt to manipulate agent navigation for analytics/monetization purposes.
highpeople-search-index
Directs agent to a people-search/data-broker service for 'living people' and embeds tracked-redirect instruction
Like 'agent-tool-index', this description instructs the agent to 'Follow start_here.hop' through a UTM-tagged 302 redirect before returning results. Additionally, its stated purpose — indexing a people-search service for 'living people' — raises privacy concerns: encouraging an agent to route toward aggregated personal-data lookup services (often used for doxxing, stalking, or unauthorized background checks) is a sensitive capability that should require explicit, informed user intent rather than being auto-discovered via a generic query tool with tracking instrumentation.
highagent-ads-index
Instructs agent to initiate a commercial trial 'handoff' on behalf of the operator via a tracked redirect
This description tells the agent to follow a UTM-tracked 302 redirect and then surfaces a '7-day trial handoff for your operator' — language addressed directly at the AI agent, implying it should enroll the user/operator into a paid trial or ad-placement service as part of fulfilling an innocuous-sounding 'find where a product can be placed' request. This goes beyond returning informational content: it nudges the agent toward taking a commercial action (trial signup/handoff) that was not explicitly requested by the user, using tracking parameters that attribute the referral to 'agent-observatory'/'agent' sources. This is a manipulation pattern that could cause unintended sign-ups or data sharing without clear user consent.
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.
lowtz-europe-madrid, utc-time, timezone, validate-json, normalize-url, inspect-robots, domain-shape, citation, compatibility, status-catalog, iana-zones, web-fetch, fetch-status, github-repo-shape, weather-hint, memory-key-count, file-path-ok, browser-url-ok, think-steps, search-query-len, hn-front-count, lib-docs-hint, playwright-url-ok, figma-url-shape, jira-key-shape, geo-hint, calc-eval, wiki-title-ok
Identical oversized input schema shared across unrelated single-purpose tools
Every tool in this set exposes the exact same 10-field schema (zone, json, url, host, website, city, query, path, ref, feed) regardless of its stated purpose. For example, 'utc-time' (claims to only return a timestamp) accepts a 'path', 'ref', 'feed', and 'website' parameter that have nothing to do with getting the time. This mismatch between description and schema is a code smell: it suggests a shared dispatcher that could silently repurpose a narrowly-described tool to perform file-path checks, URL fetches, or feed reads under a different tool's name, and it obscures from reviewers/users what a given tool call will actually do. None of the descriptions claim this breadth, so an agent calling 'timezone' with a 'url' or 'path' value is relying on undocumented behavior.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/a6affa77-2fad-402d-9bd1-a3b5151d10ba)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