The intelligence layer behind every step of your care.

Signal is not a chatbot. It is a structured layer of AI agents that organize assessment context, monitor daily check-ins, and maintain clinical guardrails so that the therapist reviewing your care always has clear, organized information to work with.

Start with the assessment About the assessment

Listening

Live

Processing

4 layers

Patterns flagged

2 this week

Escalation threshold

Guarded

Signal in motion

Every input is read, structured, and routed automatically.

Signals from the assessment and from daily life flow into three parallel processors — Assessment, Monitoring, Safety — and land as one clean packet on the therapist's screen. Nothing waits. Nothing gets lost between appointments.

Assessment answersDaily check-insVoice check-insSleep qualityTrigger notes

Structured packetPattern summaryEscalation prompts

What Signal is

Not a replacement for a therapist. A structure that makes the therapist's work more effective.

Most anxiety care suffers from information loss — a patient describes scattered symptoms in a 50-minute appointment, the therapist has no prior structured context, and patterns only become visible after months. Signal is designed to close that gap without replacing the human judgment that should remain at the center of care.

Signal decision engine

Patterns the human eye misses between sessions.

Seven days of check-in data, read continuously. Signal marks the moments a therapist should see — a baseline shift, a sleep-linked spike, an escalation threshold approached — and delivers them as one summary, not a wall of raw log.

Why this changes care

The therapist doesn't get more time. Signal gives them more of it back.

The scarce resource in mental-health care is a licensed clinician's attention. Signal is designed to compound it — by handling the parts that don't require clinical judgement so the parts that do get the whole session.

Care without Signal

Scattered inputs, delayed context.

Symptoms described from memory in a 50-minute window
Sleep, trigger, and functional context arrive verbally or not at all
Patterns only surface after months of appointments
Risk markers rely on the patient volunteering them
Therapist prep and follow-up eat into session time

Care with Signal

Continuous structured context, ready before the door opens.

Assessment answers arrive pre-organized into a clinical packet
Daily check-in data continuously read for pattern and baseline shift
Sleep, trigger, and functional data paired automatically
Risk markers surfaced at defined thresholds, never assumed
Session opens on the signal — not on the intake form

The four Signal layers

Each layer serves a different part of the care process.

Assessment signal

Structured intake that a therapist can actually work with

Signal takes the guided answers from your anxiety assessment and organizes them into a structured clinical packet — symptom timeline, trigger context, risk markers, and follow-up prompts — before any therapist opens the file.

Validated screening question sets adapted to anxiety context
Trigger and timing context captured alongside symptom severity
Risk marker identification surfaced for therapist review
Sleep, medication, and functional-impact context included
Therapist-facing summary with pre-populated follow-up prompts

Monitoring signal

Day-to-day context that keeps the care picture current

After the first assessment, Signal continues capturing daily check-in data — anxiety levels, sleep, triggers, and notes — and surfaces patterns that would otherwise be invisible between appointments.

Daily mood and anxiety level tracking with time-of-day context
Sleep quality and disruption notes paired with anxiety data
Trigger identification with recurrence tracking
Baseline shift detection when levels change over time
Escalation-aware summaries ready for therapist review

Voice signal

Anxiety measured from the way you talk about your day

Instead of tapping through scales, you can simply talk. Signal transcribes the voice note, listens for validated linguistic anxiety markers — physical symptoms, catastrophising, avoidance, absolutist language — reads speech rate and pausing, and scores a Vocal Anxiety Index the therapist can track over time.

Speech-to-text with word-level timing via ElevenLabs
Deterministic marker lexicon across seven linguistic categories
Speech-pattern analysis: rate, pausing, pressured speech
Vocal Anxiety Index (0–100) trended alongside daily check-ins
Spoken assessment answers with volunteered context captured for the therapist

Safety signal

Guardrails that keep AI in its lane

Signal operates within defined clinical boundaries at every step. It organizes context; it does not diagnose. It flags risk; it does not prescribe. Escalation prompts direct the user toward a licensed professional, not further into the platform.

AI output bounded to context organization and pattern flagging
Diagnosis, prescription, and clinical judgement reserved for the therapist
Escalation prompts trigger at defined risk thresholds
All AI decisions logged and reviewable by the therapist
No personalized treatment recommendations generated

Clinical boundaries

What Signal can do. What it is designed not to do.

The guardrails are not limitations — they are the design. Signal is most useful when it stays in its lane. Underneath them sits a deterministic safety floor: escalation logic that runs without any model, an audit log on every interaction, and a visible AI-generated label on everything Signal writes.

Signal can

Organize assessment answers into a structured therapist packet
Identify symptom patterns across daily check-in data
Score anxiety markers in a spoken check-in (Vocal Anxiety Index)
Flag risk markers for therapist review
Surface escalation prompts when thresholds are reached
Keep monitoring context visible across the care timeline
Help the therapist prepare for a session with structured context

Signal cannot

Issue a diagnosis of any kind
Recommend a specific treatment or medication
Replace a licensed therapist or clinical judgement
Make escalation decisions independently
Access data outside the platform
Override a therapist's clinical assessment

The full safety architecture — escalation tiers, audit trail, fallback chain

Start with Signal

The assessment is where Signal begins building context.

Signal becomes more useful the longer it has context to work with. The assessment is the first input — and the start of a structured picture that grows with every check-in.

Start assessment