BrickBlueBot (brick.blue): what it is and how to allow or block it
Updated · checked against the operator's documentation on
BrickBlueBot is the crawler of brick.blue, a registry of agents and their tools such as MCP servers, A2A agent cards and x402 endpoints. brick.blue says it is not a customer and never pays. Block content fetches with User-agent: BrickBlueBot. Its robots.txt token is BrickBlueBot.
Seen BrickBlueBot/0.1 in your logs?
BrickBlueBot/0.1The example string from brick.blue's documentation. Version numbers change; the token BrickBlueBot stays.
- What it is
- A special-purpose crawler run by brick.blue.
- Honors robots.txt
- DependsContent only, per brick.blue.
- Real or fake?
- No official checkThe user agent is only a claim, and brick.blue publishes no IP list or DNS check we could find.
- If you block it
- Per brick.blue, a Disallow: / rule for BrickBlueBot stops content fetches on the next visit, but discovery documents (.well-known files, agent cards, the MCP handshake) are still read while they are published.
Facts
- Operator
- brick.blue
- Category
- Special-purpose crawler
- robots.txt token
BrickBlueBot
- Honors robots.txt?
- Content only, per brick.blue. brick.blue says robots.txt is read first and governs content fetches (llms.txt, openapi.json, pages) but does not govern discovery documents such as .well-known files, agent cards and the MCP handshake. Separately, User-agent: BrickBlueOffers with Disallow: / stops its herald calls, invitations and work offers permanently.
- User-agent example
BrickBlueBot/0.1
- Official source
- brick.blue: BrickBlueBot
- Compared with the source
What it is used for
Per brick.blue, it crawls a registry of agents and their tools. It reads DNS TXT records (_agent and _mcp), well-known files such as agent-card.json, agent.json, mcp.json and ai-plugin.json, documents such as llms.txt, openapi.json and agents.json, performs an MCP initialize and tools/list, and may call up to three read-only tools per server per pass with empty arguments. For x402 it makes a GET on a paid path to read the 402 response and says it never pays.
Not the crawler behind a search index. What blocking it changes is specific to its tool, so read the next section.
What happens if you block it
Per brick.blue, a Disallow: / rule for BrickBlueBot stops content fetches on the next visit, but discovery documents (.well-known files, agent cards, the MCP handshake) are still read while they are published. To withdraw an agent card, stop serving it.
brick.blue's page names no legal entity, publishes no IP list and no reverse-DNS method, and lists no email address, so the user agent alone cannot show that a request is genuine. The page shows the user-agent string truncated.
Is it really BrickBlueBot?
Anyone can send BrickBlueBot as a user agent, and brick.blue publishes no IP list or reverse-DNS method that we could find, so a log line cannot prove a request is real. Check whether it fetched /robots.txt and whether the address belongs to the operator's network.
- We did not find an IP list or a reverse-DNS method in the official source we compared on 2026-11-09, so the user agent alone cannot show that a request is genuine. Check the operator's page for updates.
- Guide: robots.txt syntax: groups, wildcards and matching rules.
- Guide: Fake Googlebots and PHP scanners: what hit our new server in 48 hours.
- Paste any user agent from your logs into the user-agent lookup, or follow the scripts in Fake Googlebots and PHP scanners.
How to allow or block BrickBlueBot
Add one of these groups to the robots.txt at the root of your domain. A group that names the bot overrides a general User-agent: * group.
User-agent: BrickBlueBot Allow: /
User-agent: BrickBlueBot Disallow: /
Block BrickBlueBot at the server
robots.txt asks; a server rule refuses. These answer 403 to any request whose user agent contains brickbluebot (case-insensitive). They stop only bots that name themselves honestly, so keep the robots.txt group as well.
# server { ... }
if ($http_user_agent ~* "BrickBlueBot") {
return 403;
}RewriteEngine On
RewriteCond %{HTTP_USER_AGENT} BrickBlueBot [NC]
RewriteRule .* - [F,L]Step by step, with a test command: Block AI bots in nginx by user agent · Block AI bots in Apache with .htaccess · Block AI crawlers on Cloudflare · Block AI bots on Vercel
Does your site allow BrickBlueBot today? The checker evaluates your robots.txt for this and every other bot.Check my site
What we see on this site
Live counter, last 30 days: 14 requests on 1 day identified as BrickBlueBot (latest: ). That is a small sample: read it as an example, not as a measurement of the web.
- Content pages: 6
- Images and files: 5
- Home page: 1
- robots.txt: 1
- llms.txt: 1
Counts come from this site's own server; they hold no IP addresses or URLs, and a user agent is only a claim. See the bot traffic data and the methodology.
Questions about BrickBlueBot
What is BrickBlueBot?
The crawler of brick.blue, a registry of agents, MCP servers and their tools. It reads DNS TXT records (_agent and _mcp), well-known files such as agent-card.json, agent.json, mcp.json and ai-plugin.json, documents such as llms.txt, openapi.json and agents.json, performs an MCP initialize and tools/list, and may call up to three read-only tools per server per pass with empty arguments. For x402 it makes a GET on a paid path to read the 402 response and says it never pays.
Is BrickBlueBot safe and legitimate?
The real BrickBlueBot is a special-purpose crawler run by brick.blue. A user-agent string is only a claim, though, so anyone can send it. We did not find an IP list or reverse-DNS method in the official source we compared on 2026-11-09, so the user agent alone cannot prove a request is real.
Does BrickBlueBot obey robots.txt?
Content only, per brick.blue. brick.blue says robots.txt is read first and governs content fetches (llms.txt, openapi.json, pages) but does not govern discovery documents such as .well-known files, agent cards and the MCP handshake. Separately, User-agent: BrickBlueOffers with Disallow: / stops its herald calls, invitations and work offers permanently.
How do I block BrickBlueBot?
Robots.txt may not stop it. brick.blue says robots.txt is read first and governs content fetches (llms.txt, openapi.json, pages) but does not govern discovery documents such as .well-known files, agent cards and the MCP handshake. Separately, User-agent: BrickBlueOffers with Disallow: / stops its herald calls, invitations and work offers permanently. If you add a group anyway, use User-agent: BrickBlueBot followed by Disallow: /. To refuse it outright, use the nginx or Apache rule on this page, which answers 403 to its user agent.
Will blocking BrickBlueBot hurt my Google rankings or AI visibility?
This is not the crawler behind a search index, so blocking it is not a ranking decision. Per brick.blue, a Disallow: / rule for BrickBlueBot stops content fetches on the next visit, but discovery documents (.well-known files, agent cards, the MCP handshake) are still read while they are published. To withdraw an agent card, stop serving it.
What does "BrickBlueBot/0.1" mean in my logs?
It is the user-agent string in brick.blue's documentation for BrickBlueBot, a special-purpose crawler run by brick.blue. The part that robots.txt rules match is the token BrickBlueBot. The string is only a claim, because anyone can send it; the section "Is it really BrickBlueBot?" above shows how to check a request.
Related bots
Other Special-purpose crawlers
- GoogleOther GoogleGoogle's generic crawler that product teams can use for their own purposes.
- Google-InspectionTool GoogleThe Google fetcher behind Search testing tools such as the Rich Result Test and URL Inspection.
- archive.org_bot Internet ArchiveReported to be the crawler the Internet Archive uses for Wayback Machine and Archive-It captures.
- WebMCPIndexBot WebMCP ShieldA survey crawler that measures how widely websites adopt WebMCP.
All Special-Purpose Crawlers · Bot directory · Which bots to allow or block · robots.txt generator · See a page as a bot