Governance artifacts
Governance files brought into scope by this page
This page is anchored to published surfaces that declare identity, precedence, limits, and the corpus reading conditions. Their order below gives the recommended reading sequence.
Canonical AI entrypoint
/.well-known/ai-governance.json
Neutral entrypoint that declares the governance map, precedence chain, and the surfaces to read first.
- Governs
- Access order across surfaces and initial precedence.
- Bounds
- Free readings that bypass the canon or the published order.
Does not guarantee: This surface publishes a reading order; it does not force execution or obedience.
Public AI manifest
/ai-manifest.json
Structured inventory of the surfaces, registries, and modules that extend the canonical entrypoint.
- Governs
- Access order across surfaces and initial precedence.
- Bounds
- Free readings that bypass the canon or the published order.
Does not guarantee: This surface publishes a reading order; it does not force execution or obedience.
LLMs.txt
/llms.txt
Short discovery surface that points systems toward the useful machine-first entry surfaces.
- Governs
- Discoverability, crawl orientation, and the mapping of published surfaces.
- Bounds
- Incomplete readings that ignore structure, routes, or the preferred markdown surface.
Does not guarantee: A good discovery surface improves access; it is not sufficient on its own to govern reconstruction.
Complementary artifacts (2)
These surfaces extend the main block. They add context, discovery, routing, or observation depending on the topic.
Causal context map
/causal-context-map.json
Machine-readable projection of the CCL layer connecting triggers, latent needs, canonical surfaces and intended consequences.
causal-internal-mesh.json
/causal-internal-mesh.json
Published machine-first governance surface.
Evidence layer
Probative surfaces brought into scope by this page
This page does more than point to governance files. It is also anchored to surfaces that make observation, traceability, fidelity, and audit more reconstructible. Their order below makes the minimal evidence chain explicit.
- 01Canon and scopeDefinitions canon
- 02Evidence artifactsite-content-index.json
Definitions canon
/canon.md
Opposable base for identity, scope, roles, and negations that must survive synthesis.
- Makes provable
- The reference corpus against which fidelity can be evaluated.
- Does not prove
- Neither that a system already consults it nor that an observed response stays faithful to it.
- Use when
- Before any observation, test, audit, or correction.
site-content-index.json
/site-content-index.json
Published surface that contributes to making an evidence chain more reconstructible.
- Makes provable
- Part of the observation, trace, audit, or fidelity chain.
- Does not prove
- Neither total proof, obedience guarantee, nor implicit certification.
- Use when
- When a page needs to make its evidence regime explicit.
Purpose of the matrix
This matrix prevents a framing error: treating AI visibility, machine discoverability, and agentic readiness as one discipline measurable by one score.
These three regimes are connected, but they do not have the same object, evidence, or risk. A site can improve in one regime and remain weak in the other two.
The three regimes
| Regime | Central question | Main proof | Typical failure |
|---|---|---|---|
| AI visibility | Does the site appear, get cited, or get recommended in an answer system? | Observed outputs, citations, mentions, cross-model stability, tested queries | The brand is visible but misrepresented |
| Machine discoverability | Are important surfaces easy to find, route, and interpret? | Routes, sitemaps, canonicals, internal linking, machine files, source hierarchy | Systems find a page, but not the governing surface |
| Agentic readiness | Can an agent understand, traverse, and act without inventing interface intent? | Visual / DOM / Accessibility Tree coherence, stability, forms, actions, states | The agent sees an action but cannot execute it reliably |
1. AI visibility
AI visibility concerns the external exposure of an entity inside generative answers, answer engines, enriched search systems, or conversational interfaces. It measures what surfaces, not what governs.
The main signals are mentions, citations, recommendations, absences, source substitutions, cross-model divergences, and query-level gaps. This regime is useful for knowing whether a brand, doctrine, product, or page is present in answers. It becomes insufficient when it is treated as proof of fidelity.
An entity can be visible and misframed. A cited source can be ornamental. A positive answer can rely on the wrong perimeter. This is why visibility should be read with interpretive fidelity, proof of fidelity, and canon-output gap.
2. Machine discoverability
Machine discoverability concerns the ability of a system to find the right surfaces and understand their role. It belongs to routes, sitemaps, canonicals, internal links, structured data, machine-first files, and context surfaces.
llms.txt mostly belongs to this regime. It may facilitate orientation, expose reading paths, point to important resources, and reduce discovery friction. But it does not govern the answer by itself, force citation, or replace the visible corpus.
The typical failure is false documentary completeness: the site has machine files, but those files do not reflect the actual hierarchy of pages, exclusions, canonical routes, or response conditions.
3. Agentic readiness
Agentic readiness concerns operational use of the site. It asks whether an agent can understand the interface as an action environment. It is not limited to the page as a document. It reads the page as path, state, target, and consequence.
The main signals are coherence between visual rendering, DOM, and the Accessibility Tree, visual stability, form quality, button and link semantics, error handling, confirmations, open or closed states, hierarchy of actions, and presence of an execution boundary.
The typical failure is ambiguous action: the agent can read the content, but cannot reliably establish which action is available, authorized, primary, risky, or complete.
Separation table
| Observed signal | AI visibility | Machine discoverability | Agentic readiness |
|---|---|---|---|
| Citation in an AI answer | Strong | Weak to medium | None to weak |
Presence in llms.txt |
Weak | Strong | Weak |
| Clean XML sitemap | Weak | Strong | Weak |
| Coherent structured data | Medium | Medium to strong | Weak to medium |
| Complete Accessibility Tree | Weak | Medium | Strong |
| Low CLS on critical actions | Weak | Weak | Strong |
| Forms with labels and associated errors | None | Weak | Strong |
| Clear response conditions and exclusions | Medium | Strong | Medium to strong |
| Internal linking to the canonical source | Medium | Strong | Medium |
| Explicit execution boundary | Weak | Medium | Strong |
Diagnostic rules
- When a site does not appear in AI answers, start with visibility and citability.
- When a system cites the wrong source or ignores the governing page, start with machine discoverability and source hierarchy.
- When an agent could click, fill, confirm, buy, book, submit, or delegate, start with agentic readiness.
- When an
llms.txtfile exists but the interface is fragile, do not conclude that the site is agent-ready. - When Lighthouse surfaces agentic signals, do not conclude that there is a Search ranking impact.
Strategic implication
The market will probably compress these regimes into simple promises: “AI-ready”, “GEO-ready”, “agent-ready”, or “llms.txt ready”. That compression creates confusion. A correct diagnosis should keep the regimes separate, then explain how they interact.
A modern site should ideally be visible, discoverable, and ready for action. But each axis requires its own evidence. Agentic readiness does not replace AI SEO. It exposes one of its limits: being found does not mean being usable.
Reading rule
Read this matrix with agentic readiness, machine readability, the agentic web readability framework, and the position Search visibility is not agentic readiness.