Anthropic MCP Goes GA, Forcing Immediate Agent-Facing Infrastructure Decisions
Anthropic's Model Context Protocol reached general availability with expanded third-party server support, making structured agent-facing endpoints a baseline expectation rather than an experiment. Businesses without MCP-compatible data surfaces risk being invisible to the growing class of Claude-powered agents making autonomous vendor recommendations.
This Week's Signals
Anthropic Formally Launches MCP General Availability with Multi-Server Orchestration Support
Why it matters for your score
MCP GA means Claude agents can now query multiple registered business endpoints in a single session, compare offerings, and surface recommendations without a human in the loop. Businesses that have published an MCP server — even a minimal one exposing pricing, availability, and service descriptions — gain a structural advantage over those relying solely on web crawls. If your business data isn't reachable via a machine-queryable endpoint, you are not in the consideration set for autonomous Claude agents.
Perplexity Tightens Citation Logic, Favoring Pages with Explicit Structured Pricing and FAQ Schema
Why it matters for your score
Perplexity's answer engine has shifted toward pages where prices, service tiers, and FAQs are marked up in JSON-LD, making those fields directly parseable rather than inferred from prose. Businesses without PriceSpecification and FAQPage schema are seeing citation rates drop even when their content ranks well in traditional search. Adding schema isn't a future-proofing task anymore — it's the immediate difference between being cited or being skipped.
llms.txt Adoption Crosses 50,000 Verified Domains as Tooling Ecosystem Matures
Why it matters for your score
The llms.txt directory has crossed 50,000 verified publisher files, a threshold that signals the format has moved from early adopter to expected infrastructure — similar to where sitemaps were in 2006. Major AI crawlers including GPTBot and ClaudeBot have confirmed they ingest llms.txt as a priority signal for understanding site structure and preferred content surfaces. Businesses without a filed llms.txt are now the visible minority, not the majority, in competitive verticals like SaaS, legal, and professional services.
What to do this week
- 1
Publish or audit your llms.txt file this week — the 50,000-domain milestone means your competitors in any established vertical are likely already filed. Use Faro's /tools/llms-txt generator to produce a correctly structured file that surfaces your core services, pricing page, and contact endpoint in the format GPTBot and ClaudeBot prioritize.
- 2
Add PriceSpecification and FAQPage JSON-LD to your highest-traffic service pages before Perplexity's next crawl cycle. Use Faro's /tools/ai-schema tool to generate validated schema blocks tailored to your business type — paste directly into your page head without touching your CMS templates.
- 3
Run your AI Readiness Scan at /tools/ai-readiness-scan to get a scored gap report across MCP exposure, llms.txt status, schema coverage, and robots.txt crawler permissions — then prioritize the MCP endpoint section, since Anthropic's GA release means agent-facing infrastructure is now the highest-leverage gap for most SMB and agency accounts.