Glossary
Canonical definitions of terms used throughout the AXAG specification.
A
Action — A discrete operation that an agent can perform on an entity. Defined by an intent, entity, and action type.
Action Type — Classification of an action as read, write, delete or navigate.
Affordance — A property of an interface element that communicates what actions are possible. In AXAG, affordances are declared via axag-* attributes rather than visual cues.
Agent — An AI system that performs tasks autonomously on behalf of a user by interpreting semantic contracts and invoking tools.
Agent Experience (AX) — The discipline of designing interfaces and systems that are optimized for autonomous agent consumption, distinct from User Experience (UX).
Annotation — An axag-* HTML attribute added to a UI element to declare its semantic intent, parameters, and constraints.
Approval — A gate that requires authorization from a designated role before an action can execute. Stronger than confirmation.
AXAG — Agent Experience Accessibility Guidelines. The specification that defines how to annotate web interfaces for agent consumption.
C
Confirmation — A gate that requires explicit user acknowledgment before a high-risk action executes.
Conformance Level — A tier of AXAG compliance: basic (required fields), intermediate (risk and safety metadata), full (scope and declared conditions as well). The legacy names A, AA, AAA map onto these.
Constraint — A validation rule on a parameter (type, range, enum, format, pattern).
D
Determinism — The property that an action produces predictable, consistent results given the same inputs and preconditions.
E
Entity — A domain object that actions operate on (e.g., product, cart, user, ticket).
I
Idempotent — An operation that produces the same result when called multiple times with the same parameters.
Intent — The canonical identifier for an action, following the entity.verb pattern in snake_case (e.g., product.search, cart.add_item).
Inheritance — A container's axag-scope, axag-tenant-boundary and axag-required-roles apply to the annotations inside it. See Context Inheritance.
M
Manifest — See Semantic Manifest.
Macro — The axag shorthand attribute that packs identity and safety fields into one string, e.g. axag="write:user.deactivate!critical?approval". Expands to the longhand attributes.
MCP — Model Context Protocol. A standard for AI agent tool integration. AXAG-generated tools target MCP format.
H
Harvesting — Reading parameters from the markup — a form's controls, their names, labels, types and HTML constraints — rather than declaring them on the annotation. See Schema Harvesting.
P
Postcondition — A condition guaranteed to be true after successful execution of an action.
Precondition — A condition that MUST be true before an action can execute.
R
Risk Level — Classification of an action's potential impact: none, low, medium, high, critical.
S
Safety Boundary — The set of constraints (confirmation, approval, scope, tenant boundary) that protect against unintended agent actions.
Scope — The data access boundary for an action: public, user, tenant, global.
Semantic Contract — The machine-readable agreement between a UI and its agent consumers, defined by AXAG annotations and the Semantic Manifest.
Semantic Manifest — A JSON document that aggregates all AXAG annotations from a site into a discoverable, machine-readable contract.
Side Effect — An external consequence of executing an action (e.g., email sent, payment processed, inventory updated).
T
Tenant Boundary — The isolation enforcement level for multi-tenant operations: strict (no cross-tenant access) or relaxed.
Tool — A callable function generated from a Semantic Manifest action, consumable by AI agents via MCP or similar protocols.
Tool Generation — The process of converting Semantic Manifest actions into MCP-compatible tool definitions.
W
WebMCP — The browser API (document.modelContext) through which a page registers tools with an agent running in the browser. See Agent Runtime.