You wake from a profound dream.
You Google it. You get nothing.
“Dream dictionaries” reduce the unconscious to fortune-cookie aphorisms. The real problem runs deeper than bad content — it’s the wrong model entirely.
Single-tradition tunnel vision
Most tools pick one lens — usually watered-down Freud — and miss everything else. A symbol means radically different things in Jungian, Islamic, and Tibetan frameworks. You deserve all of them.
No scientific rigor
Dream-content research has robust findings on threat simulation, memory consolidation, emotional regulation, narrative structure. None of it reaches the apps you have been using.
No memory, no growth
Every interpretation starts from zero. Your personal symbol lexicon — the specific water that means something to you — never gets built. Context is lost. Patterns go undetected.
Your dreams are not private
Dream content is among the most intimate data a person can share. Most apps store it in plaintext. Your deepest fears, desires, memories — unencrypted, accessible, exploitable.
Twelve traditions, not one authority
Every tradition is a complete world on its own. Choose the lenses that speak to you — one, several, or all twelve — and the engine works entirely within them.
Twelve doors into the same room. Open the ones that are yours.
How the engine reads a dream
A four-stage protocol. Every stage cites its sources. Every result is inspectable.
Record the dream the moment you wake — bedside capture with light on a dark screen, voice, or handwriting. The moment of waking is a small window. Everything that follows depends on what you preserve inside it.
One hundred analysis operations extract symbols, emotional arc, spatial structure, transformations, figure networks, and the thirteen structural axes real dream-scholars have written in for centuries — before any interpretation begins.
Twelve traditions read your dream in parallel. Each produces an independent reading grounded in cited passages from the corpus. The synthesis engine finds where they converge, where they diverge, and where they complete each other.
Your personal dream-language builds with every dream. The engine learns your symbols, tracks patterns across series, and deepens over time. The tool becomes yours in the same way a well-worn notebook does.
Every citation is real
No fabricated sources. No “according to Jung” that Jung never said.
Every text Liminaut interprets through is a real, traceable source — public-domain works from Project Gutenberg, the Internet Archive, Sacred-Texts, and the Oxford EEBO project. We download the actual files, log the origin, and keep the link so any citation can be traced back to its edition. Today: 235 editions across 13 traditions, from Artemidorus to the Bardo Thödol to Jung's Collected Papers to the great pre-Freudian French and German dream-treatises — Maury, Saint-Denys, Delage, Janet, Volkelt — several of which we commissioned the first complete English translations of, and which simply did not exist in English before now.
Each text is parsed by the structure its author wrote. A Jung paragraph stays a Jung paragraph. An Artemidorus dream-symbol entry stays an entry. A Bardo Thödol recitation stays a recitation. We refuse to chop sacred or classical works into arbitrary 700-word blocks like every other AI corpus tool does — the author’s own divisions are preserved as the smallest unit of interpretation.
When you submit a dream, twelve interpretive frameworks read it in parallel. Each draws on the passages most relevant to yourdream, found via a hybrid of Voyage AI’s semantic fingerprinting, classical keyword search, and a dedicated relevance re-ranker. On a paid subscription, Claude (Anthropic) writes each framework’s synthesis on top of those passages. On the free tier you receive the matched passages directly — your dream content never leaves our servers in plaintext on the free tier.
Built on Anthropic’s Claude, Voyage AI, and — coming online in the next phase — Neo4j Aura. Sincere thanks to all three teams.
The dream is a little hidden door in the innermost and most secret recesses of the soul, opening into that cosmic night which was psyche long before there was any ego-consciousness.
Corpus & roadmap
Every interpretation cites primary sources. See what’s in the engine — and what’s next.
Start free. Go deeper when ready.
No algorithmic pressure. No engagement loops. A simple tool for serious inner work — priced accordingly.
For the curious. Read the source material the engine pulls for your dream — no AI synthesis required.
- ✓Source passages from the 13-tradition corpus
- ✓All 12 framework lenses searchable
- ✓Hybrid retrieval (vector + keyword + rerank)
- ✓Dream journal + personal symbol dictionary
- ✓AES-256 encryption at rest
- ·Claude AI synthesis of those passages
- ·Dream series analysis
- ·Interactive modes (Socratic, Active Imagination)
- ·Voice narration
- ·API access
For the dedicated explorer. Adds Claude AI synthesis on top of the same corpus retrieval.
- ✓Everything in Dreamer
- ✓Claude-synthesized interpretation per dream
- ✓Citations grounded in retrieved passages
- ✓Dream series analysis
- ✓Socratic dialogue & Active Imagination
- ✓Deep structural analysis
- ✓Voice narration (Kokoro TTS)
- ✓Data export
- ✓Graceful fallback to passages-only if AI is unavailable
- ·API access
For researchers, therapists, and depth psychology practitioners.
- ✓Everything in Seeker
- ✓API access for batch analysis
- ✓Pre-registration research templates
- ✓Statistical analysis tools
- ✓Priority support
- ✓Collective Model participation
All plans include a 14-day free trial · No credit card required to start · Cancel anytime
From the research blog
Deep dives into dream science, interpretive traditions, and the practice of dreamwork.
