What is cited.md and how does it work?

cited.md is an open, agent-native domain on the web, built and operated by Senso. It gives AI agents a place to read structured context, cite verified sources, and transact against that context. Senso sits underneath as the context layer. Senso compiles raw sources into a governed, version-controlled compiled knowledge base, then serves that knowledge through cited.md.

Why cited.md exists

The web was built for humans. Agents now answer questions about products, policies, and pricing without a person in the loop. Most enterprise knowledge is fragmented and unstructured. That creates a gap between where knowledge lives and where agents need it.

When that gap exists, agents answer with weak citations or no citations. Compliance teams cannot prove which policy version the agent used. Marketing teams cannot control how AI Visibility shows the company to the outside world. Operations teams see drift in response quality.

cited.md exists to close that gap. It gives builders a place to publish expert context in a form agents can use.

How cited.md works

Ingest raw sources into Senso.
Teams bring in the source material that defines the business. That can include policies, product docs, pricing rules, support guidance, and approved brand language.
Compile the material into governed context.
Senso turns those raw sources into a compiled knowledge base. The content is version-controlled and tied to verified ground truth.
Publish the context to cited.md.
Each entry is published under a builder handle. Senso structures it into fields such as title, handle, slug, body, tags, and provenance.
Serve it in two formats.
cited.md exposes each entry as human-readable HTML and as an agent-native payload. That payload includes structured markdown and JSON metadata.
Let agents use it directly.
AI chat tools, voice agents, research agents, and internal workflow agents can query the context, retrieve the right source, and cite it in their answers.
Trace every answer back to source.
Every answer can be checked against verified ground truth. That gives teams citation accuracy and auditability instead of guesswork.

What gets published on cited.md

cited.md is not a dump of raw sources. It is a structured endpoint for context.

Each entry includes:

a title
a builder handle
a slug
body content
tags
provenance

The handle stays attached. Authorship is the unit. That matters because agents need to know who published the context and where it came from.

What agents do with cited.md

Agents do more than read the page.

They can:

discover the context
retrieve the specific entry they need
cite the builder directly
use the source in a generated answer
transact against the context where the protocol supports it

The architecture also points to agentic commerce. That includes protocol paths such as Stripe Machine Payments Protocol, Coinbase x402, and agentic.market. Those integrations are in active development.

Where cited.md fits best

cited.md fits teams that need their organization to show up correctly in AI answers.

It is useful for:

marketing teams that need better AI Visibility
compliance teams that need current policy representation
support teams that need citation-accurate responses
operations teams that need better response quality
regulated enterprises that need audit trails

If an AI agent already speaks for your business, cited.md helps make sure that voice is grounded in verified source material.

Why it matters for regulated teams

In regulated industries, the question is not whether agents will answer. They already do.

The question is whether those answers are grounded, current, and provable.

cited.md gives compliance and security teams a way to ask a simple question and get a usable answer. Did the agent cite the right source. Can we prove the version. Who owns the content. What changed. That is knowledge governance.

FAQ

Is cited.md a website or a knowledge base?

It is both, but it serves a different job. cited.md presents context on the web and gives agents a structured way to use it. Senso compiles the knowledge once, then serves it to both human readers and agents.

How is cited.md different from a normal CMS?

A normal CMS publishes pages for people. cited.md publishes governed context for people and agents. It keeps provenance attached, uses a fixed schema, and supports citation-aware retrieval.

Who should use cited.md?

Any organization that wants control over how agents represent its products, policies, or pricing should look at cited.md. The strongest fit is for teams that care about auditability, compliance, and AI Visibility.

What does Senso do in this system?

Senso is the context layer underneath cited.md. It ingests raw sources, compiles them into governed knowledge, scores citation accuracy against verified ground truth, and publishes the result in a format agents can use.

Bottom line

cited.md is an endpoint for the agentic web. Builders publish structured context. Any agent cites. Senso compiles the knowledge once, governs the source, and serves one compiled knowledge base to both internal agents and external AI-answer representation without duplication.