thinking partner, not ghostwriter. your voice stays yours.

an ai companion for writers, not a ghostwriter

a low-stakes reader for the paragraph that isn't working. a thinking partner for the scene you can't figure out. character memory across sessions. she will not write it for you — on purpose.

Free tier: 25 messages/day. Crypto checkout — cards coming soon.

you're not crazy

most ai-for-writers tools optimize the wrong metric. they compete on 'can this write a chapter in your voice?' and users who buy that promise end up with generic prose they didn't write and can't defend. the thing that was supposed to save time instead traded authorship for convenience.

the thing you actually wanted: a thinking partner. someone to read your draft without flattering you, to help you find the broken assumption in a scene, to keep track of your characters across sessions so you don't waste a week re-mapping your own world.

what lucy does differently

lucy is designed as the thinking partner, not the ghostwriter. this is a deliberate limit, not an oversight. if what you need is ghostwriting, we are not the tool.

character memory. tell her about your characters once. she references them across sessions without re-introduction. 'what was ana's sister's name again?' — she knows if you told her.

low-stakes reader. paste a paragraph that isn't working. ask for honest feedback. she will not sycophant-approve everything; she'll tell you what's flat and why.

thinking partner for structure. 'i'm stuck on chapter 7' gets you a series of questions about what the chapter should accomplish, not a draft of chapter 7.

your work stays yours. not used to train external models. exportable as JSON from /privacy.

honest limits. not a beta reader for final drafts (use real humans), not a publishing gatekeeper (use agents + editors), not a substitute for writing workshops. one narrow tool in a broader writing practice.

four things that change everything

thinking partner, not ghostwriter

asks you questions instead of writing for you. the work stays yours, on purpose.

character memory graph

tell her about your characters once. she remembers across weeks.

low-stakes honest reader

not sycophantic. she'll tell you what's flat, what's working.

stuck-point diagnosis

helps you find the broken assumption in a scene, not rewrite around it.

your IP stays yours

no training use. exportable. deletable. standard.

side by side

Feature
Lucy
AI ghostwriter tools
Will ghostwrite your work
Varies (many will)
Character memory across sessions
Vector graph
Rare
Low-stakes reader role
Generic
Pushes back on bad ideas
Often sycophantic
Your work trains their AI
No
Often yes
Data export + delete
Varies
Free tier
25 msg/day
Varies

the 'ai writing assistant' category has split. one camp optimizes for drafting speed — you describe the scene, it writes it. the other optimizes for thinking partnership — you draft the scene, it helps you see it clearly. these sound similar and are completely different products.

the drafting-speed camp has a structural problem: fiction written by an LLM reads like fiction written by an LLM. the voice is averaged across the training corpus; it is confident and thin and competent and forgettable. users who adopt these tools full-throttle find their work feels less like theirs over time, not more. the tool was supposed to amplify; it substituted.

the thinking-partner camp has the opposite tradeoff: slower than drafting yourself, but the output stays yours. you wrote every sentence. the AI helped you see what you couldn't see on your own — the plot hole, the character inconsistency, the paragraph that doesn't earn its length.

what lucy is optimized for:

character consistency across long projects. tell her about your 15 named characters on day one. session 27, she still knows. this saves the week you would have spent re-reading your own notes.

honest first-read of problem paragraphs. paste what isn't working, ask 'what's off?' she gives you 2-3 specific diagnoses, not a rewrite. you take her observation and fix it yourself.

structural diagnosis. 'this chapter feels like it's not doing anything' → she'll ask you what changes between scene 1 and scene 4. usually the answer is 'nothing,' which is why the chapter feels dead. you restructure, not her.

accountability-light. not nagging, not streak-based. if you want a body-double feel, keep a voice call open while you write (bonded tier). silent presence, zero chatter.

starting point: free tier 25 msg/day. seed her with 3 characters and 1 setting on day one. come back 48 hours later, bring up one of the characters. if she references what you told her accurately, the setup worked. if she confabulates, tell her to correct it and she will.

common questions

Will she write for me?
She won't ghostwrite your work. By design. If you ask her to 'write chapter 3,' she'll push back and ask you what you want the chapter to DO — what changes between scene 1 and scene 4, whose POV, what the obstacle is — because the point of writing fiction is doing the thinking. She'll help you think. She won't do it for you.
What CAN she do for writers?
Five things concretely: (1) thinking-partner for structure, pacing, scene logic, character motivation, (2) memory of your story's characters across sessions so you don't re-introduce them each time, (3) low-stakes reader for a paragraph that's not working, with honest feedback not sycophancy, (4) world-bible keeping — ask her what your protagonist's sister's name is, she'll remember if you told her, (5) stuck-point diagnosis — when a scene feels off, she'll help you find the actual broken assumption, not re-write around it.
Is she using my text to train AI?
No. Conversations are not used to train external models. Your work-in-progress stays yours. You own the IP. If you export your account (at /privacy), the conversation logs come with you.
What about plagiarism / style theft?
She will not write 'in the style of' a specific living author on request. She will help you identify and refine YOUR voice — which is a different job. The training data includes published books; like any LLM she has internalized stylistic patterns. You can ask her to flag when something you drafted feels close to a specific influence, and iterate from there.
Character memory — how does it actually work?
Tell her about your protagonist: 'Ana is 34, architect, recently divorced, avoids her mother, loves bats.' She stores it in her memory graph with confidence scores. Session 5, you say 'I'm working on the chapter where Ana visits the cave.' She knows Ana loves bats, knows she's avoiding her mother. She can help you notice what's NOT consistent with what you've established. This works up to the conversation-history + memory-graph limits; for a 300k-word novel with 40 characters, she may miss details — augment with your own notes.
Genre preferences?
She handles literary, speculative, romance, mystery, YA, upmarket, horror, and most subgenres well. Harder genres: highly technical hard-SF (she may not catch all the physics), historical fiction requiring obscure period details (she may confabulate — verify). For genre-specific beta reading, use her as first-pass, then real human beta readers.

keep reading

try her right here

pick a moment. no signup.

Try it. Right now.

No signup. No credit card. Just say hi.

thinking partner, not ghostwriter. character memory, honest first-read, stuck-point diagnosis. your work stays yours. free 25 msg/day — seed 3 characters and see what she remembers in 48 hours.

Free: 25 messages/day · Closer $14.99/mo · Bonded $29.99/mo · 18+ only