Ad Radar MCP server
dev.ad-radar/ad-radarad-radar.devstreamable-httpv1.2.0Scanned
Search winning Meta ads with $50k+ in estimated spend, read hooks and scripts, and draft briefs.
6 tools: 6 only read.
- Reads private data · 0
- Reads text others write · 1
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Ad Radar MCP server can do
The Ad Radar MCP server is published by ad-radar.dev, a domain the MCP Registry verified. It has 6 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
1 tool brings in text other people can write (get_ad).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 14, 2026, version 1.2.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
6 tools
Only reads 6Show 6 read-only tools
advertiser_hooksReturns the hooks, hook mechanisms and angles an advertiser uses most, counted over all of its ads in the library.No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Returns the hooks, hook mechanisms and angles an advertiser uses most, counted over all of its ads in the library. advertiserId is the advertiser id shown in search_ads rows.
Annotations the server declares
- title: Advertiser hooks
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_adReturns the full record of one ad: headline, body, CTA, full transcript, metrics, spendEstimate (an ESTIMATE based on engagement, not real spend; spendLowerBound = it is a minimum,…Reads text others writeNo findings
Why it is in this group
- The name "get_ad" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "return" and "post": it reads posts, which other people can write.
Description, as the server wrote it Returns the full record of one ad: headline, body, CTA, full transcript, metrics, spendEstimate (an ESTIMATE based on engagement, not real spend; spendLowerBound = it is a minimum, "at least"), delivery status, winnerObservation (when we first observed it above $50k), media (public URLs), links (post, Ad Library, landing page, board), tags (the same dimension and slug search_ads shows), publisher platforms and landing domain. The transcript is in English: when the ad runs in another language (e.g. Spanish), transcript is the English translation, translatedFrom names the language and transcriptOriginal holds the original script. Text fields are null when they aren't available for that ad; the links always point to the original.
Annotations the server declares
- title: Ad details
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_swipe_promptReturns a prompt template built from one ad, written for the user to run on their own product.No findings
Why it is in this group
- The name "get_swipe_prompt" says "get": a word for reading.
Description, as the server wrote it Returns a prompt template built from one ad, written for the user to run on their own product. The result is JSON with the template text in the template field. kind='swipe' is a template that maps the ad's beats, checks the fit with the user's product, drafts a full version with the same structure and adds alternative hooks. kind='analysis' is a template that breaks down why the ad works (audience, funnel, psychology, weak spots) and how its structure applies to a business. product and niche are written into the template; without product the template keeps a placeholder for it.
Annotations the server declares
- title: Swipe prompt
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_taxonomyLists the taxonomy dimensions of the Ad Radar library (awareness, angle, hook_pattern, vformat, pformat, landing...) with their valid values.No findings
Why it is in this group
- The name "list_taxonomy" says "list": a word for reading.
Description, as the server wrote it Lists the taxonomy dimensions of the Ad Radar library (awareness, angle, hook_pattern, vformat, pformat, landing...) with their valid values. These slugs are the values the tags filter of search_ads accepts.
Annotations the server declares
- title: Taxonomy
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
search_adsSearches the Ad Radar library of Facebook and Instagram ads from the Meta Ad Library (every ad has at least $50k estimated spend), with the same filters as the Ad Radar board.No findings
Why it is in this group
- The name "search_ads" says "search": a word for reading.
Description, as the server wrote it Searches the Ad Radar library of Facebook and Instagram ads from the Meta Ad Library (every ad has at least $50k estimated spend), with the same filters as the Ad Radar board. By default it returns only active ads, sorted by score, as compact rows. Each row has an adKey, which get_ad and get_swipe_prompt accept. Valid tag dimensions and slugs are the ones list_taxonomy returns. headline and body are null when the ad's text isn't available (the tags still describe the ad); in rows they are cut at 140 and 280 characters, while get_ad returns them whole. Every other value in a row matches what get_ad returns for that ad. spendEstimate is an estimate based on engagement, not reported spend; spendLowerBound true means the value is a minimum ("at least").
Annotations the server declares
- title: Search ads
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
winning_patternsReturns, for each taxonomy value, how many times more often it appears among winning ads than on average (lift), with adoption (ads, advertisers) and example ads; the same data as …No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Returns, for each taxonomy value, how many times more often it appears among winning ads than on average (lift), with adoption (ads, advertisers) and example ads; the same data as the 'What to do now' block of the Ad Radar analytics page. doThis lists the values above average, lower the values below. exampleAdKeys are adKeys that get_ad accepts.
Annotations the server declares
- title: Winning patterns
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write1
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:51 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://app.ad-radar.dev/.well-known/oauth-protected-resource/api/mcp
- Authorization servers
- https://app.ad-radar.dev
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- yes
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.2.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http ad-radar https://app.ad-radar.dev/api/mcpclaude mcp add --transport http ad-radar https://app.ad-radar.dev/api/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__ad-radar__search_ads",
"mcp__ad-radar__get_ad",
"mcp__ad-radar__get_swipe_prompt",
"mcp__ad-radar__list_taxonomy",
"mcp__ad-radar__advertiser_hooks",
"mcp__ad-radar__winning_patterns"
],
"ask": [],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as ad-radar. Review it before you use it.
Capability tags behind 1 rule
- mcp__ad-radar__get_ad untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/dev.ad-radar/ad-radar/verify?hash=aa08a04dc47c8645dc376b08f77ade954aa64a3af33026639846c0209b23d3aa" | jq -e '.changed == false'Pinned hash aa08a04dc47c, from Nov 14, 2026, 23:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/dev.ad-radar/ad-radar/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/dev.ad-radar/ad-radar)Source: MCP Registry. Synced Oct 21, 2026, 18:11 UTC.
Ad Radar MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Ad Radar MCP server safe?
- No scan can promise that. What its definitions show: 6 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the Ad Radar MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Ad Radar MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Ad Radar MCP server have?
- 6 tools: 6 read. search_ads, get_ad, get_swipe_prompt, list_taxonomy, advertiser_hooks and winning_patterns.
Does the Ad Radar MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Ad Radar MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Ad Radar MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 14, 2026.