AI Agent Context Platforms
What is cited.md and how does it work?
7 min read
AI agents already answer questions about products, policies, and pricing. The problem is not whether they can respond. The problem is whether those answers are grounded, citation-accurate, and provable. cited.md is Senso’s answer to that gap. It is an open, agent-native domain where builders publish structured context and agents cite, retrieve, and transact against it.
Quick answer
cited.md is an endpoint for the agentic web. Senso compiles an organization’s raw sources into a governed, version-controlled compiled knowledge base, then serves that knowledge as both human-readable HTML and an agent-native payload. Each entry keeps the builder handle, title, slug, tags, and provenance attached, so agents can trace answers back to verified ground truth.
For teams that care about AI Visibility, compliance, and auditability, that matters because agents are already representing the business whether humans review the output or not.
What cited.md is
cited.md is an open, agent-native domain where experts publish context and agents cite it. It is built and operated by Senso.
The idea is simple. The web was built for humans. Agents need their own endpoint. They need a source of expert-verified context they can cite, learn from, retrieve, and transact against.
cited.md does that by making context:
- Open to the web
- Structured at the source
- Attributed to the builder who wrote it
- Usable by agents without extra interpretation
- Traceable back to provenance and verified ground truth
The core rule is also simple. Any builder publishes. Any agent cites.
How cited.md works
cited.md works in three stages.
1. Senso compiles the knowledge
Senso ingests raw sources and compiles them into a governed knowledge base. That matters because enterprise knowledge is usually fragmented and unstructured. Agents often have to infer across stale pages, policy docs, and inconsistent internal materials.
Senso turns that sprawl into a single compiled knowledge base that can be governed and version controlled.
2. Builders publish structured context
Each entry is published under a builder’s handle and structured into Senso’s schema:
- Title
- Handle
- Slug
- Body
- Tags
- Provenance
That structure keeps authorship attached to the content. It also gives agents a clean way to read, cite, and reuse the context without guessing where it came from.
3. cited.md serves context in two formats
Senso serves each entry as:
- Human-readable HTML
- An agent-native payload with structured markdown plus JSON metadata
That dual format matters. People can read it directly. Agents can query it and use it without losing context.
The handle stays attached in both formats. Authorship is the unit.
Why that matters for enterprise teams
Most enterprises do not have a knowledge problem. They have a governance problem.
Agents are already telling customers, employees, and partners what the organization believes about its products, policies, and pricing. If that information is fragmented, old, or uncited, the organization can be misrepresented in public and exposed in regulated workflows.
cited.md helps with that in four ways:
- It gives agents a verified source to cite
- It reduces drift between raw sources and agent responses
- It creates a traceable path from answer to provenance
- It lets compliance and operations teams see what agents are saying
For regulated industries, that audit trail is the point. A CISO should be able to ask whether an agent cited the current policy and whether the organization can prove it. cited.md is built for that question.
What problems cited.md solves
cited.md is designed for teams that need grounded answers, not generic responses.
It helps when you need to solve:
- Fragmented knowledge across teams and systems
- Uncited or stale agent responses
- Inconsistent external representation in AI results
- Duplicate work between internal agents and public-facing AI answers
- Slow review cycles when no one can trace the source of an answer
It also helps marketing and compliance teams control how AI models represent the organization externally. That is where AI Visibility becomes operational, not theoretical.
How cited.md fits into the agentic web
The broader idea behind cited.md is that the web is becoming agent-readable, not just human-readable.
In that model:
- Builders publish structured context
- Agents discover it
- Agents retrieve it
- Agents cite it
- Agents transact against it
The docs also point to emerging agentic commerce flows through protocols such as Stripe Machine Payments Protocol, Coinbase x402, and agentic.market. Those integrations are described as active development, but the direction is clear. Context is becoming something agents can fetch and act on, not just read.
How cited.md differs from a standard CMS or knowledge base
| Layer | Standard CMS | Internal knowledge base | cited.md |
|---|---|---|---|
| Primary audience | Humans | Employees | Agents and humans |
| Structure | Pages and posts | Internal docs and files | Structured context with handle, slug, tags, and provenance |
| Citation | Manual | Limited | Built in |
| Auditability | Weak | Varies | Stronger, because each entry stays tied to verified ground truth |
| External AI use | Indirect | Usually none | Direct |
The key difference is not just format. It is governance. cited.md is built so agents can cite a specific source, and teams can prove where that answer came from.
Proof points from Senso deployments
Senso reports outcomes that show why governed context matters:
- 60% narrative control in 4 weeks
- 0% to 31% share of voice in 90 days
- 90%+ response quality
- 5x reduction in wait times
Those results are about what changes when organizations stop leaving agent responses to chance and start grounding them in verified context.
Who should care most
cited.md matters most for teams that need both representation and proof.
Marketing and brand teams
Use it when you need control over how AI systems describe your products, policies, and positioning.
Compliance teams
Use it when you need traceability, version control, and a record of what agents said and why.
CISOs and IT leaders
Use it when you need citation accuracy, policy alignment, and auditability across internal agents.
Operations leaders
Use it when response quality and wait time matter, and drift creates work for staff.
FAQs
What is the simplest way to describe cited.md?
cited.md is an open, agent-native domain where organizations publish structured context so agents can cite it, retrieve it, and use it as verified ground truth.
Is cited.md a website or a knowledge base?
It is both, but neither label fully captures it. It is a web endpoint for agent-readable context, backed by Senso’s compiled and governed knowledge layer.
How does an agent use cited.md?
An agent reads the structured entry, follows the attached handle and provenance, and cites the source when it answers. That reduces guesswork and makes answers traceable.
Why does provenance matter here?
Provenance shows where the context came from. That matters when a policy changes, a price changes, or a regulated answer needs proof.
Is cited.md only for external AI answers?
No. The same compiled knowledge can support internal workflow agents and external AI-answer representation. That avoids duplication and keeps the source of truth consistent.
Bottom line
cited.md is Senso’s endpoint for the agentic web. It gives organizations a way to publish structured context that agents can cite, retrieve, and transact against. The value is not just reach. It is governance. Every answer stays tied to a handle, a source, and verified ground truth.
If your organization is already using agents, the question is not whether they represent you. The question is whether you can prove what they said and where it came from.