P01 · Before meaning
pattern_id: P04_expectation_setting
Expectation Setting
Before the system acts, it states what is about to happen, how long it may take, and what it will and will not do.
emote · patterns
Trust moments & patterns
emote patterns are behavioral contracts for moments where trust is thin. They specify what the system must do, what it must not do, and how behavior stays reviewable through token sets, guardrails, and safe fallbacks. This becomes especially important for AI-driven and automated systems.
What’s inside every pattern
Every pattern follows the same reviewable structure, so teams can evaluate behavior like a spec, not a vibe.
Pattern contract
Canonical spec: promises, rules, and the token set.
Pattern in practice
A before/after showing the same user message, different behavior.
Live demo
What the user sees, plus UI surfaced and a JSON reference.
Behind the scenes
Shared flow: Detect → Orient → Boundaries → Preflight → Fallbacks.
Browse behavioral tokens
Tokens are reusable pieces. Patterns are the contracts that compose them.
Trust pattern sequence
These patterns map to a small set of recurring trust moments. They’re designed to be composed: the same token may support multiple patterns, while each pattern anchors to a specific moment in time.
P02 · Meaning breaks
pattern_id: P01_ambiguity_detection
Ambiguity Detection
When intent is unclear, the system pauses, clarifies, and avoids guessing.
P03 · Meaning restored
pattern_id: P03_interpretive_support
Interpretive Support
The system clarifies options and context without steering outcomes, reducing cognitive load without pressure.
P04 · Agency check
pattern_id: P02_consent_confirmation
Consent Confirmation
Before sensitive or boundary-crossing actions, the system verifies permission and restates what will happen.
P05 · Trust repair
pattern_id: P03_repair_apology
Repair & Apology
When the system causes harm or confusion, it names it, takes responsibility, and offers a concrete path to repair.
P06 · Trust continuity
pattern_id: P06_state_reorientation
State Reorientation
After a rupture or state change, the system re-anchors what changed, what remains true, and what agency the user still holds.
Notes: “Irreversible actions,” “user vulnerability,” “escalation,” and “identity & dignity” are treated as signals, modifiers, or constraints that shape how these patterns run, not standalone patterns themselves.
How to use these patterns
emote patterns are meant to sit alongside your existing design system, research practice, and engineering stack. Use them wherever intent, risk, or emotion make behavior sensitive, not routine.
In UX specs
Call out patterns and tokens in flows and requirements. Example: “If login intent is unclear, apply P01 Ambiguity Detection with behavior.clarify_before_action before showing troubleshooting steps.”
In prompts & agents
Treat patterns as reusable behavior contracts in a prompt library or policy layer. Reference the pattern ID and apply the token set, for example:
Use pattern P01_ambiguity_detection. Apply tokens: - behavior.pause_when_uncertain - behavior.clarify_before_action - behavior.name_risk_transparently - behavior.delay_irreversible_actions
In product decisions
Use patterns to align PMs, designers, researchers, and engineers on what “good behavior” means. When debating a feature, ask: “Which trust moment is this, and are we honoring the emote pattern for it?”
View token reference
Tokens are designed to be reused across patterns and domains.
emote patterns are intentionally small. Combine them with your domain-specific rules, safety systems, and visual tokens. Think of them as the behavioral layer that keeps systems aligned with people, especially when the stakes are higher than a happy path.