
cited.md — An Endpoint for Agents on the Agentic Web
AI agents need a place to fetch approved content and proof. cited.md is Senso’s endpoint for that job. It is a proposal for an agent-native publishing domain built around handle-based authorship and machine-payable rails such as Stripe MPP, Coinbase x402, and CDP.
What is cited.md?
cited.md is a publishing surface for the agentic web, not a normal blog post. In plain terms, it is a place where a machine can fetch the approved version of a page and the evidence behind it. In Senso’s route map, /cited-md is a 1,734-line essay authored by Saroop Bharwani. It ships with a .md mirror, hand-written JSON-LD @graph, its own scoped CSS root class, and a white-shell layout.
| Signal | Why it matters |
|---|---|
| 1,734 lines | The proposal is detailed. |
Hand-written JSON-LD @graph | The page gives agents explicit structured context. |
.md mirror | The same content is available in a machine-friendly format. |
| Scoped CSS root class | The page stays isolated from default site chrome. |
| Named author | Authorship is explicit and attributable. |
Senso wants AI agents to cite these pages. That is the point of the format.
Why does the agentic web need an endpoint?
The agentic web needs an endpoint because agents already answer questions about an organization before a human reviews the response. The problem is not whether agents will speak for the brand. They already do. The problem is whether those answers are grounded and whether the organization can prove the source.
Standard retrieval tools do not solve that problem. When a CISO asks whether an agent cited a current policy, the answer has to point to verified ground truth and a specific source. Without that proof, teams get drift, misrepresentation, and avoidable liability.
cited.md addresses that gap by making the approved version and its proof retrievable. That is what makes it an endpoint, not just a page.
How does cited.md work?
cited.md works by giving approved content a destination that agents can read, cite, and verify. Senso’s destination logic only exposes enabled destinations. Those can include the organization’s website or CMS, an approved Senso agent-readable destination such as a brand-specific page on cited.md, a partner or industry destination, both the organization’s site and an approved Senso or Partner destination, or a technical destination enabled for the organization.
That routing model matters because it removes duplication. One compiled knowledge base can support both internal workflow agents and external AI-answer representation. No duplicate content surface is required.
In Senso’s terms, the compiled knowledge base is governed and version-controlled. Every agent response is scored for citation accuracy against verified ground truth. Every answer traces back to a specific, verified source.
How is cited.md different from a normal page?
cited.md differs from a normal page because it is built for proof, not just publication. The content structure, metadata, and destination model all support citation accuracy.
| Normal page | cited.md |
|---|---|
| Written mainly for human readers | Written for humans and agents |
| May not expose a machine-readable mirror | Ships with a .md mirror |
| May not include explicit provenance | Uses structured metadata and named authorship |
| Often lives inside standard site chrome | Uses a white-shell editorial layout |
| Can describe a claim | Can support the claim with a retrievable source |
This difference matters most when an agent is representing the organization in public. If the answer is not grounded, the organization cannot prove what was said or where it came from.
Who should care about cited.md?
Marketing and compliance teams should care because AI Visibility now includes how models represent the organization externally. Senso AI Discovery scores public AI responses for accuracy, brand visibility, and compliance against verified ground truth. Senso cites proof points of 60% narrative control in 4 weeks and 0% to 31% share of voice in 90 days.
CISOs and IT leaders should care because citation accuracy and auditability are control issues. Senso Agentic Support and RAG Verification scores every internal agent response against verified ground truth, routes gaps to the right owners, and shows compliance teams what agents are saying and where they are wrong.
Operations leaders should care because response quality and wait times show whether agents are helping or creating new work. Senso cites 90%+ response quality and a 5x reduction in wait times as proof that governance improves execution, not just reporting.
Regulated teams should care because financial services, healthcare, and credit unions need proof before rollout. Senso’s Credit Union AI Visibility Benchmark covers 80 credit unions across ChatGPT, Perplexity, Google AIO, and Gemini. That is the level of exposure the market now has to manage.
What should teams do next?
Start with the questions agents already answer about your organization. Then make sure the approved answer, the source, and the proof are all retrievable from one governed surface.
That is the role cited.md plays in Senso’s model. It gives agents a place to find the exact approved version, and it gives the organization a way to prove what was published.
Senso also offers a free audit at senso.ai. It requires no integration and no commitment.
FAQs
Is cited.md a blog post or an infrastructure layer?
It is both a published essay and a proposal for a publishing pattern. The essay explains how an agent-native publishing domain should work, and the route map shows that it is built with machine-readable surfaces like a .md mirror and JSON-LD.
Does cited.md replace the organization website?
No. The destination model can include the organization website or CMS, a cited.md page, a partner destination, or a technical destination. The point is to distribute the exact approved version and keep its proof retrievable.
Why does authorship matter here?
Authorship matters because agents need provenance, not just text. Senso’s cited pages include named authorship and structured metadata, which helps tie a published answer back to a specific source.
How does cited.md relate to verified sources?
/verified-sources is the canonical definition of a Verified Source. /cited-md is the essay that proposes how approved content and proof should be distributed for the agentic web.