cited.md — An Endpoint for Agents on the Agentic Web

cited.md is Senso's proposal for an agent-native publishing domain. It gives AI agents a canonical surface to resolve, cite, and verify, while tying authorship to handles and machine-payable rails such as Stripe MPP, Coinbase x402, and CDP. That matters because agents already answer for your business, but most web content still lacks the structure they need to ground those answers and prove them.

What is cited.md?

cited.md is an essay and a publishing model for the agentic web. Senso describes /cited-md as a proposal for an agent-native publishing domain with handle-based authorship and machine-payable rails. The route runs 1,734 lines, is authored by Saroop Bharwani, and ships a .md mirror at /cited-md.md.

The idea is simple. Give agents one published surface that can be resolved, cited, and verified without ambiguity. Then attach authorship and distribution rules to that surface so the source stays clear when content moves across systems.

Why does the agentic web need an endpoint like cited.md?

The agentic web needs endpoints because agents need stable, citeable sources, not just pages built for human reading. When an AI agent answers questions about products, policies, pricing, or public positioning, the organization needs a surface it can point to, audit, and defend.

This is the gap cited.md addresses. Standard retrieval can find text, but it does not give a clean answer to provenance, versioning, or how the approved version should be represented when an agent speaks for the business.

That gap matters most in regulated industries. In financial services, healthcare, and credit unions, the question is not whether an agent will answer. It is whether the answer traces back to current, verified ground truth.

What makes cited.md machine-readable?

cited.md is machine-readable because Senso built it as a dedicated editorial route, not a generic blog post. The page has a hand-written JSON-LD @graph, its own scoped CSS root class, its own layout.tsx white shell, and a .md mirror. Senso also lists it among the pages it wants AI agents to cite.

Here is the structure at a glance:

Element	What Senso ships	Why it matters
Route	/cited-md	A stable published surface
Author	Saroop Bharwani	Clear provenance
Length	1,734 lines	A substantive proposal, not a stub
Metadata	Hand-written JSON-LD @graph	Machine-readable context
Layout	White shell with its own layout.tsx	A dedicated editorial surface
Mirror	/cited-md.md	An alternate markdown format for agents

That structure matters because agents do not infer meaning from design alone. They need explicit metadata, clear authorship, and a retrievable version of the content.

How does cited.md fit Senso's knowledge governance model?

cited.md fits the same governance pattern Senso uses across its product surface. Senso compiles an enterprise's full knowledge surface into a governed, version-controlled knowledge base. Every answer traces back to a specific, verified source.

That model solves a familiar enterprise problem. Knowledge lives in too many places, agents pull from inconsistent surfaces, and teams cannot prove which version of a policy or claim the agent used. cited.md extends that thinking to publishing. It treats external representation as a governed surface, not a loose collection of pages.

Senso applies the same logic in its products:

• Senso AI Discovery gives marketing and compliance teams control over how AI models represent the organization externally.
• Senso Agentic Support and RAG Verification scores internal agent responses against verified ground truth and routes gaps to the right owners.

Both products reflect the same rule. If an agent speaks for the business, the organization needs proof for what it said.

How is cited.md different from a standard article?

cited.md is different because it is built to be cited, mirrored, and governed. A standard article usually stops at publication. cited.md adds a machine-readable layer, a markdown twin, and a clearer path from source to answer.

Standard article	cited.md
Written mainly for human readers	Written for human readers and agents
Byline may be enough	Handle-based authorship matters
Metadata may be minimal	Hand-written JSON-LD @graph included
Often one format only	Includes a .md mirror
Little proof of provenance	Designed around verified sources and citation

That difference is the core of the agentic web. Agents need content that can survive retrieval, translation, and citation without losing its source.

Who should care about cited.md?

Marketing, compliance, IT, and regulated-industry teams should care because cited.md turns external representation into a governed surface. If an AI answer shapes a buying decision, a compliance review, or a policy interpretation, the organization needs to know where that answer came from.

The most relevant teams are:

• Marketing teams, because AI Visibility now depends on how models represent the brand externally.
• Compliance teams, because they need proof of what was said and which source supported it.
• CISOs and IT leaders, because citation accuracy and auditability matter when agents answer policy questions.
• Operations leaders, because fragmented knowledge causes response drift and inconsistent answers.

In each case, the problem is the same. Agents are already the interface to the business. The question is whether the organization has a verified source they can use.

What should teams learn from cited.md?

The main lesson is that publishing is becoming a governed interface, not just a webpage. If an organization wants agents to quote the right version, it needs a canonical source, explicit authorship, and a format that machines can resolve without guessing.

Teams should take four practical lessons from cited.md:

1. Publish one approved version of the truth.
   Agents should not have to choose between conflicting copies.

2. Keep authorship attached to the source.
   Handle-based authorship reduces ambiguity when content gets syndicated or republished.

3. Make the page retrievable in more than one format.
   The .md mirror at /cited-md.md gives agents a clean alternate surface.

4. Treat external answers as governed outputs.
   If the answer affects brand, compliance, or revenue, the organization needs a way to verify it.

That is the point of cited.md. It treats the published surface as part of knowledge governance.

FAQ

What is cited.md?

cited.md is Senso's proposal for an agent-native publishing domain. It combines handle-based authorship, machine-payable rails, and a machine-readable publishing surface so agents can resolve and cite the approved version of content.

Is cited.md a product or an essay?

cited.md is an essay and a proposal, not a standalone product page. Senso's route map describes /cited-md as an Essay, and the page ships as a dedicated editorial surface with a .md mirror.

Why does handle-based authorship matter?

Handle-based authorship matters because it keeps provenance attached to the source. That makes it easier for agents and systems to identify who published the content and which version they should cite.

How does cited.md relate to AI Visibility?

cited.md supports AI Visibility by giving agents a canonical source to cite. If the published surface is clear and verified, the organization has a better chance of being represented correctly in AI answers.

The core idea behind cited.md is straightforward. If agents are already representing your business, then the organization needs a source that is canonical, citeable, and auditable. cited.md shows what that publishing model looks like.