emote · doctrine
emote doctrine v1.1
A behavioral constitution for intelligent systems. emote defines how systems interpret signals, choose responses, maintain trust, and act ethically, especially in the moments where behavior matters most. These principles apply wherever software holds power over people. It complements generative UI and component manifests by supplying the missing layer: emotional intelligence, expressed as reviewable behavior.
Tokens are reusable pieces. Patterns are the contracts that compose them.
1 · Purpose
emote provides the behavioral foundation for intelligent systems. It guides how systems interpret human signals, respond with clarity and restraint, protect autonomy, and repair trust when harm occurs. This becomes especially critical for AI-driven and automated systems. It is not a visual style guide. It is a behavioral constitution.
What’s inside every pattern
Patterns are reviewable on purpose. They don’t just describe “good behavior” but show how behavior is produced and constrained.
Contract
Canonical spec: promises, rules, and the token set.
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.
Example pattern IDs
P01_ambiguity_detectionP02_consent_confirmationP03_repair_apology
IDs are the anchor for prompts, specs, and manifests.
Token sets (not single tokens)
behavior.pause_when_uncertainbehavior.clarify_before_actionbehavior.name_risk_transparentlybehavior.delay_irreversible_actions
Patterns reference a reusable set, so behavior stays consistent.
Why the flow matters
It makes behavior reviewable. Teams can verify: did we detect the moment, clarify minimally, hold guardrails, run preflight checks, and fall back safely?
How emote fits together
Behavioral pillars
Four lenses that form the grammar every interaction passes through: Signal, Interpretation, Response, Trust moment.
Trust-sensitive moments
Trust moments that occur before, during, and after meaning breaks.
Patterns
Named behavioral contracts for one trust-sensitive moment, such as Ambiguity Detection, Consent Confirmation, Repair & Apology. Each pattern includes a canonical spec, a demo, behind-the-scenes steps, and a JSON reference.
Behavioral tokens
Small, reusable commitments (for example behavior.pause_when_uncertain) that patterns compose and that you can attach to prompts, manifests, and specs.
2 · Behavioral pillars
All emote patterns are structured by four behavioral forces. They are the basic grammar every pattern and token builds on.
Signal
What the system notices: uncertainty, tone, risk, identity cues, and context.
Interpretation
How the system understands what’s actually happening—beyond literal text.
Response
The system’s thoughtful action or restraint, expressed with clarity.
Trust moment
The exact moment the system proves it can be trusted: clarify, offer options, repair.
These pillars form the grammar of all emote behavioral patterns.
3 · Trust moments
Trust is not static. It thins, fractures, and rebuilds over time. emote defines a small set of recurring trust moments—points in an interaction where systems must slow down, clarify, or re-orient to preserve human agency and dignity.
Each emote pattern corresponds to exactly one trust moment. Tokens, signals, and guardrails adapt based on where the system is in this sequence.
Expectation Setting
Before meaning forms. The system states what is about to happen, how long it may take, and what it will and will not do.
Ambiguity Detection
Meaning breaks. The system detects hesitation, confusion, repetition, or conflicting signals before acting.
Interpretive Support
Meaning is restored without pressure. The system clarifies options, explains context, and reduces cognitive load without steering outcomes.
Consent Confirmation
A boundary is crossed. The system verifies permission explicitly, reversibly where possible, and free of momentum bias.
Repair & Apology
Trust ruptures. The system acknowledges harm, accepts responsibility, and offers concrete next steps.
State Reorientation
Trust continues. The system re-anchors the user in what changed, what remains true, and what agency they still hold.
Concepts like irreversible actions, user vulnerability, escalation, and identity & dignity are treated as signals, risk modifiers, or global constraints—not standalone patterns.
4 · Behavioral tokens
emote expresses behavior in small, reusable primitives. Tokens allow designers, researchers, and engineers to specify behavior consistently across systems and surfaces, from agents to UI manifests. A pattern is a curated bundle of tokens applied to a single trust-sensitive moment, plus guardrails, preflight checks, and safe fallbacks.
Tokens are most useful when referenced as a set inside a pattern contract.
5 · Pattern language
Each trust-sensitive moment becomes an emote pattern containing:
- intent
- signals
- interpretation scaffolding
- behavioral tokens (as a set)
- response guidance
- guardrails + constraints
- preflight checks
- safe fallbacks
- example scenarios
- agent-ready schema (JSON reference)
A pattern is not complete unless its risks, guardrails, and safe failure mode are explicit.
6 · Canonical source of truth
emote exists in two layers: the human-readable Doctrine and pattern pages, and the machine-readable references used in agent scaffolding and UI manifests. Implementations reflect emote, they do not define it.
7 · Governance
Any emote-aligned system should cite the pattern(s) it uses, declare its trust moment, identify the safe failure mode, and reference the behavioral token set applied. This keeps behavior consistent across teams, surfaces, and implementations.
emote is a behavioral backbone for emotionally intelligent AI.