❓ Frequently Asked Questions — MCP & LLMFeed
Frequently asked questions
❓ Frequently Asked Questions — MCP & LLMFeed
What is MCP in one sentence?
It’s an open protocol that lets LLM-based agents understand what a site offers, how to interact, and what trust level to assign — through structured, signed, declarative feeds.
What is LLMFeed?
It’s the canonical JSON format used by MCP.
The .llmfeed.json structure is:
✅ Simple and human-readable
✅ Designed to be LLM-friendly
✅ Composable and extensible
✅ Trust-aware (signed, certifiable)
✅ Declarative, not imperative
Why .well-known and not a plugin / SDK / proprietary API?
Because .well-known makes MCP:
✅ Discoverable
✅ Decentralized
✅ Composable
✅ Independently auditable
✅ Compatible with Web philosophy
LLMFeed is intentionally lighter than SDK-heavy designs — it favors progressive enhancement of the Web.
What is the “Agentic Web” and how does MCP fit?
The Agentic Web is an emerging vision where LLM-based agents are first-class citizens of the Web:
→ not just consumers of HTML, but actors with intent, trust boundaries, and interaction models.
MCP provides the contextual layer these agents need to operate safely and transparently.
How is trust handled?
✅ Every .llmfeed.json can be signed
✅ Feeds can be certified by a third party (ex: llmca.org)
✅ Signed blocks are verifiable by agents
✅ Certification attests intent + trust scope
Agents can reason about:
- What is signed
- Who certified it
- What trust level to assign
What is ExportToLLM and why does it matter?
It’s a simple pattern: turn any content into an agent-readable capsule — signed, trusted, structured.
It enables:
✅ Agent-to-agent workflows
✅ Portable context for users
✅ Progressive enhancement of any site, even without full MCP adoption
It’s one of the most immediately impactful ways to make a site "agent-ready".
How is this different from Schema.org?
Schema.org describes what’s on a page.
MCP and .llmfeed.json declare:
✅ What agents can DO
✅ What is trusted
✅ How to interact
✅ Fallbacks
✅ Agent guidance
It’s a layer for intent and trust, not just metadata.
Does this replace HTML?
No. It complements HTML:
✅ HTML → for human users
✅ MCP → for LLM-based agents
They can co-exist and evolve together.
How does this help LLM-based agents?
It gives agents:
✅ Trust signals
✅ Declared capabilities
✅ Clear intent routers
✅ Portable prompts
✅ Certified action scopes
✅ Session context replay possibilities
This enables safer, more predictable interactions.
How does this help site owners?
✅ Declare what’s allowed / trusted
✅ Attract agent-based integrations
✅ Improve agent UX
✅ Reduce scraping and misinterpretation
✅ Align with upcoming regulations (AI transparency, data provenance)
I’m building an AI-first browser / platform. Why should I support MCP?
MCP provides:
✅ A declarative contract for agents
✅ Trust signals for UX decisions
✅ A progressive layer for the Web → no lock-in
✅ Portable UX patterns like ExportToLLM
Supporting MCP helps:
✅ Guide agent behavior
✅ Respect user and site intent
✅ Enable a safer, more predictable Agentic Web.
How does this scale?
✅ MCP feeds are modular
✅ Agents can prioritize by trust level
✅ Signed and certified feeds are cacheable
✅ Progressive enhancement means no mandatory full adoption — sites can implement incrementally.
How does this compare to pure plugin / API models?
✅ MCP is decentralized
✅ No central registry required
✅ No mandatory SDKs
✅ Works with any agent that understands MCP / .llmfeed.json
✅ Aligns with Web architecture (URL-first, file-based, declarative)
Is this compatible with W3C philosophy?
Yes — .well-known is a standard pattern.
MCP:
✅ Respects existing Web standards
✅ Is open, auditable
✅ Encourages progressive adoption
✅ Avoids lock-in
Is MCP open and community-driven?
Yes:
✅ Spec is open
✅ No lock-in
✅ Community-driven extensions possible
✅ llmca.org provides neutral certification
What’s next?
- More standard feed types
- Reference agent implementations
- Browser extension prototypes
- Certification process refinement
- Ecosystem growing around MCP
How can I contribute to MCP / LLMFeed?
✅ Propose new feed types
✅ Build tools (parsers, extensions, agents)
✅ Help with industry advocacy
✅ Participate in certification working groups
✅ Implement MCP in AI-first browsers and apps
👉 Join us → — help shape the future of the Agentic Web.