What is cited.md and how does it work?

Agents are already representing your organization. They answer questions about your products, policies, and pricing without a human in the loop. The issue is whether those answers are grounded in verified ground truth, and whether you can prove where they came from. cited.md is Senso’s open, agent-native domain for that problem. Builders publish structured context. Agents cite it, retrieve it, and can transact against it.

cited.md is built and operated by Senso. It is an endpoint for the agentic web. The web was built for humans. Agents need their own endpoint.

What cited.md is

cited.md is a place on the web where experts and organizations publish context in a format agents can use.

• It is an open, agent-native domain.
• It is designed for expert-verified context.
• It lets agents cite, retrieve, learn from, and transact against published content.
• It keeps authorship attached through a builder handle.
• It serves the same content to humans and agents, but in different formats.

In plain terms, cited.md gives agents a source they can point to. It turns raw sources into structured context that can be cited and traced.

Why cited.md exists

Most enterprise knowledge is fragmented. It lives across raw sources, policy pages, docs, and internal systems. Agents do not handle that mess reliably on their own.

That creates three problems:

1. Misrepresentation. An agent can answer with stale or incomplete information.
2. No proof. A CISO or compliance leader cannot easily prove which source the agent used.
3. No control. Marketing and compliance teams cannot see how external models represent the company.

cited.md is built for that gap. It gives organizations a published, governed place for context that agents can cite.

How cited.md works

cited.md works by connecting three layers.

1. Senso compiles the knowledge

Senso ingests raw sources and compiles them into a governed, version-controlled knowledge base.

That matters because agents need more than scattered files. They need a compiled knowledge base with verified ground truth behind it.

2. Organizations publish context to cited.md

Senso’s agentic CMS lets organizations engineer context and publish it to a domain on the web, such as cited.md.

Each entry is structured into Senso’s schema:

• title
• handle
• slug
• body
• tags
• provenance

The handle stays attached. Authorship is the unit. That makes the source of each entry explicit.

3. The Context Layer serves it to agents

Senso’s Context Layer serves each entry in two formats:

• human-readable HTML
• an agent-native payload with structured markdown and JSON metadata

That means a person can read the page, and an agent can parse the same content without guessing.

4. Agents cite, retrieve, and transact

AI chat tools, voice agents, research agents, and other systems can read the published context, cite it, and use it in responses.

The published architecture also points to agentic protocols such as:

• Stripe Machine Payments Protocol
• Coinbase x402
• agentic.market

The docs describe this as an active direction for the build. The model is simple. Publish context once. Let agents use it.

5. Responses are checked against verified ground truth

Senso does not stop at publishing.

It scores each agent response against verified ground truth. That lets teams measure citation accuracy and find gaps fast.

If an answer is wrong, Senso can route the issue to the right owner. That is what makes cited.md useful for governance, not just distribution.

What makes cited.md different from a normal website

Layer	What it does	Why it matters
Senso	Compiles raw sources into governed knowledge	Creates verified ground truth
cited.md	Publishes structured context on the web	Gives agents a source they can cite
Agents	Query and cite the published context	Reduces guesswork and drift
Teams	Review citation accuracy and gaps	Improves control, auditability, and response quality

A normal website is built for people to read. cited.md is built so agents can read, cite, and act on what they find.

Who cited.md is for

cited.md matters most for teams that need control over how agents represent the organization.

Marketing and brand teams

They use cited.md to shape AI Visibility. That means they can see how public models describe the company and what needs to change.

Compliance teams

They use cited.md to confirm that answers trace back to verified sources and current policy.

CISOs and IT leaders

They use cited.md to answer a hard question. Can the organization prove what the agent said, and where it came from?

Operations leaders

They use cited.md to improve response quality and reduce wait times when agents handle frontline questions.

Regulated industries

Financial services, healthcare, and credit unions need auditability. They need citation accuracy. They need a clear record of what an agent said and why.

What outcomes Senso has reported

Senso has published the following proof points from deployments:

• 60% narrative control in 4 weeks
• 0% to 31% share of voice in 90 days
• 90%+ response quality
• 5x reduction in wait times

Those numbers matter because they show the point of cited.md. It is not just about publishing content. It is about controlling how agents represent the organization and proving that the answers are grounded.

What Senso does underneath cited.md

cited.md is part of a larger system.

Senso compiles an enterprise’s full knowledge surface into one governed compiled knowledge base. That same compiled knowledge base can support:

• internal workflow agents
• external AI-answer representation
• compliance review
• citation verification
• gap routing to the right owner

That avoids duplication. One source of truth can serve both internal and external agent use cases.

Simple example

Here is the basic flow.

1. Your team ingests raw sources.
2. Senso compiles them into governed knowledge.
3. You publish the context to cited.md under your handle.
4. Agents query that published context.
5. The response includes a traceable source.
6. Senso scores the response against verified ground truth.
7. Your team sees where the agent was right or wrong.

That is the core loop. Publish. Cite. Verify. Improve.

What cited.md is not

cited.md is not a generic file dump.

It is not a loose collection of pages with no provenance.

It is not a human-only site that agents have to guess their way through.

It is a structured endpoint for expert-verified context on the agentic web.

FAQs

Is cited.md a CMS or a domain?

It is both in practice. Senso uses an agentic CMS to publish context, and cited.md is the domain that serves that context to the web.

Does cited.md replace a knowledge base?

No. It sits on top of compiled knowledge. Senso compiles the knowledge once, then serves it in a format agents can use.

Why does provenance matter?

Because agents already represent your organization. If the answer is wrong, you need to know which source it came from and whether it matched verified ground truth.

Who controls the content on cited.md?

The builder does. Every entry is published under a handle, with structured provenance attached.

What is the main business value?

Control, auditability, and citation accuracy. That is what protects brand representation and reduces risk when agents answer for you.

If you want to see how agents represent your organization today, Senso offers a free audit at senso.ai. No integration. No commitment.