AI Coaching sidebar (interview prompts and artifacts)
AI Coaching is a virtual project in Omniscio where you sit down with Claude for a structured interview and walk away with a saved artifact — how to turn it on, choose a coaching style, work the Dashboard, and what happens to your sessions and bookmarks along the way.
On this page
What it is
AI Coaching is a virtual project in Omniscio where you sit down with Claude for a structured interview — answering questions about a topic (your week, a project decision, a goal you're chasing) and, at the end, walking away with a saved artifact (a short written summary, plan, or reflection that Claude composed from the conversation). It lives in the sidebar like any other project. Selecting it swaps the main panel for an AI-Coaching-specific layout: a sub-sidebar on the left listing your interview prompts, in-flight sessions, saved artifacts, and archived sessions, with the matching content opening in the right pane.
The point is not just to chat with Claude — it's to capture the result. Interview sessions produce artifacts you can re-open later. Think of it as guided journaling with a memory.
On a phone (mobile): AI Coaching shows a bottom tab bar — Dashboard / Sessions / Artifacts / Bookmarks / Profile — pinned above the phone's home bar. Dashboard is the interview prompt cards (finished interviews collapse to compact single-row cards — badge, title, View artifact, ⋮ — so a long completed list stays scannable); Sessions lists your coaching sessions (tap one to open its conversation, just like any other mobile session); Artifacts lists your saved artifacts (tap one to open it); Bookmarks lists your saved bookmarks (content, label, date) with a search box and, once you have labelled any, a label filter — plus the same add, label-edit and delete controls as the desktop pane; only the grouped-by-label view stays desktop-only; Profile shows your Core Profile, and lets you refresh it — the Regenerate control is icon-only on a phone, beside Copy — without waiting for the out-of-date banner to offer it (editing the profile text stays desktop-only). Every tab renders its content or an explicit empty state, so the pane is never blank; an empty Sessions or Artifacts tab says what to do next and offers Browse interviews, which opens the Dashboard. Tap a prompt's Start interview and the conversation opens; the back arrow returns you. (Archived and Prompts are under More; tapping a search result, a Context Index row or a prompt opens it right on the phone.) On a completed card, View artifact opens that artifact right on your phone — the title, the formatted text, a copy button, a compression level picker on the latest version (Full / Key Points / Summary / Micro; a level not generated yet is greyed out) and, once it has more than one saved version, a version picker (v1 … latest) — with a back arrow to return. Switching a level only changes what you read; it does not change which level feeds your other sessions. On the latest version, Edit lets you change the text and Save it; restoring an old version and deleting an artifact stay desktop-only. (Earlier builds showed a blank pane here, and the dashboard couldn't scroll past the fold on a phone — both fixed; see ai-coaching-mobile-layout-contract.md.) The ⋮ Pinned context picker works on the phone too (since 2026-10-14); Import, Bulk Import, Export, JSON and Delete all data read or write files on the computer, so they show only in the desktop app.
Where to find it
Turning it on
AI Coaching is off by default. To enable it, open Settings → AI Coach → "Enable AI Coaching" and flip the toggle on. The toggle sits inside the AI Coaching card, whose subtitle reads "Guided coaching interviews, your saved write-ups and progress tracking." The virtual project row appears in the left sidebar immediately, no relaunch needed. Toggling it back off makes the row disappear, blocks every AI Coaching IPC call with "AI Coaching is disabled. Enable it in Settings → AI Coach.", and short-circuits any already-approved cli-pending artifact mutations still queued in the dispatcher (those land as permanent failures rather than executing against a now-hidden project). The default-off behavior is enforced at three layers (IPC handlers, settings-apply symmetric ON/OFF, cli-pending dispatcher) so the toggle is the real on/off switch, not just a sidebar-visibility flag.
Coaching style
Once AI Coaching is enabled, a Coaching Style dropdown appears below the toggle in Settings → AI Coach. Three options:
- Warm (default) — affirming, patient, and gently paced. A
<coaching_tone>block with the warm guidance is appended after the shared context in CLAUDE.md, the same as for the other two styles. - Neutral — professional, balanced, and collaborative. The block tells the coach to keep the tone respectful and focused.
- Direct — efficient, straightforward, and incisive. The block tells the coach to push for specifics, prioritise thoroughness over comfort, and ask one question at a time; that one-question rule wins over any questions-per-round count elsewhere in the instructions.
The style applies to every kind of coaching session: structured interviews, free-form chats opened with + New in the coaching project, and check-ins (template check-ins and Quick Check-ins). It takes effect on the next session you start — changing it does not alter sessions that are already running. Internally, buildCoachingToneSection() in context/claudemd-builder.ts emits the block for every style (warm when no style is saved), and each session builder reads the style from settings.aiCoachingStyle.
How to use it
- Open it. Click AI Coaching in the left projects sidebar. The right side of the app splits into a sub-sidebar (left) and a content pane (right), the same shape as Quick Replies or Automations.
- Pick a section. The sub-sidebar has four sections from top to bottom:
- Dashboard — the landing card. Shows your interview prompts organized so what's next leads — an In progress tier (any live interview), then a browse-by-theme area (everything still to do), then Completed at the bottom (dimmed). A genuine first visit opens with a dismissible welcome (what this is, plus the three steps — pick a topic, have the conversation, keep what you discover); it shows only until you've engaged — one completed or in-progress interview, or a Core Profile, hides it — and because that check reads your saved history rather than a per-device flag, it won't reappear when you open Omniscio on a new phone or browser. Above the tiers a journey strip shows how far you've come (a progress bar, plus "2 of 8 complete" and the time left once you have finished your first conversation), your recommended next step, and the Core Profile milestone kept in view ("N more saved write-ups to unlock your Core Profile" until you have three saved write-ups the profile can use, then "your profile is ready"). Finishing a conversation — or unlocking your profile — marks the moment with a warm note; it only ever celebrates something genuinely new, never work that was already done. Completing every interview triggers a special graduation toast plus a persistent banner on the dashboard (dismissible); the banner stays until dismissed and persists across visits. You browse the interviews by theme: a filter bar (All · Foundation · Deep analysis · Problem solving · Work · Ongoing · Optional, each with a live count) plus a search box sit on top, and the interviews are grouped into theme sections beneath — pick a theme, or type to search, to narrow the list (a Clear filters button shows when nothing matches). Any interviews of yours that don't fit those themes land in a More group, so nothing is ever hidden. The built-in journaling and therapeutic interviews — Hard Truth Journal, What I Am Avoiding, CBT Thought Records, IFS Parts Work, ACT Values Clarification, Stoic Reflection, First Principles Analysis, and Grief, Loss and Major Life Transitions — have no theme of their own yet, so they appear in More too, each labelled simply "Interview" where other cards name their category. No interview is ever locked — every card is startable the moment you click it; a card that draws on earlier interviews simply shows a quiet "Builds on …" note (guidance, not a gate). Each available card leads with a small category label (Foundation, Work, Deep analysis, Problem solving, Ongoing, Optional) beside its colored dot, then the title, a short pitch — a one-line hook plus up to three "what you walk away with" points — and a Start interview → link whose arrow slides on hover. An In progress card swaps the dot for a green spinner (a live session, the same green a running session wears elsewhere in Omniscio) and shows just its hook above a phase-progress bar. A completed card is cleaner: just a "done" badge, the title, and a View artifact button on its face, with Redo, Reset, and — when the interview that produced it still exists — Open chat tucked into a ⋮ menu (no pitch, no category label, no "Quick check-in" shortcut). On a phone this card collapses to a compact single row — badge, title, View artifact, and the ⋮ — so a long finished list stays scannable instead of a stack of tall cards.
- Sessions — your in-flight interview sessions, grouped by status (see below). Click a row and that session opens in the right pane, exactly as if you'd opened it from the main sidebar.
- Artifacts (N) — the artifacts you've saved from past interviews, newest first. Clicking one opens the artifact editor in the right pane (described in detail below). The count appears once your artifacts have loaded; until then the header shows just the name, never a 0. Bookmarks (N) works the same way. If your artifacts can't be loaded, the list says Couldn’t load your artifacts. with a Retry button instead of claiming you have none — the same on the phone's Artifacts tab.
- Archived (N) — sessions you've archived, flat list, newest first. Click to re-open in the right pane. The count is your full archived total; if that total can't be read, it counts the rows shown, and with no rows it shows just the name.
- Start a new interview. From the Dashboard, click any prompt card — each one is titled (e.g., "Weekly review", "Project retrospective") and shows a short pitch (a one-line hook plus a few "what you walk away with" points) plus a Start interview → link at the bottom. Clicking the card instantly shows the interview's first message — an authored greeting that appears as Claude's opening turn with no waiting and no cost — then switches you into the Sessions section and selects the new row. The real Claude session doesn't start until you send your first reply; until then the row is a quiet draft you can leave and come back to. (See Opening messages below for how this deferred start works.) Nine interviews that ask about sensitive things — Health & Wellness Assessment, Somatic and Body Awareness Assessment, Marriage and Relationship Deep Analysis, Grief, Loss and Major Life Transitions, Parenting and Family Dynamics, Formative Experience Deep Dive, IFS Parts Work, CBT Thought Records and Current Life Reality Assessment — first show a short notice, This interview covers sensitive topics. It says what the interview asks about, that what you share goes to the AI service that runs your coach, and whether the write-up stays inside coaching. Nothing starts until you press Start interview; Not now closes it. The notice shows every time, Redo and the wrap-up Next button included. The short check-in versions of these interviews do not show it. When your coach asks to start one of the nine, the inbox always asks you first with its usual Start coaching interview card, even if you have switched interview approvals off.
- Run the interview. Use the right-hand pane like any other Claude session — answer the questions, hit Enter, and Claude drives the conversation. When it has gathered enough material, the interview prompt teaches Claude to save the finished document with one call to Omniscio, which answers whether the save worked, and to also show that same document inline in the chat as clean prose. The new version shows up in the Artifacts section without you doing anything. The save never stops for an approval prompt, even at the default permission level: Omniscio recognizes the coach's own save and bookmark commands, word for word, and lets only those through, and only in a coaching session; any other command still asks. If you want to nudge Claude to wrap up, hit the Save artifact now button at the top of the session pane: it sends a one-line message asking Claude to save the document (and show it) right now. The button is debounced 5 seconds and stays disabled while the coach is replying, until you have answered at least once, and when the session is paused, archived, errored, or ended. A long interview also runs in segments: each time it passes one of the topic's planned break points, a banner appears above the chat — "Segment 3 of 6 complete — take a break or continue" — with Continue to carry straight on and Pause Session to stop there and resume the same conversation later. The offer waits for you: if a break point passes while another session is in front, the banner is still there the next time you open that interview, and it goes away once you pick either button.
- Review your artifacts. Switch to the Artifacts section. Each artifact row shows a title, a version number (artifacts can be re-saved into the same slot — version increments), and a date. Click to open the artifact editor in the right pane. From there you can edit the contents, switch to historical versions, restore an old version, copy or export the markdown, or delete the artifact entirely. Each of those flows is covered in its own section below.
- Export everything. On the Dashboard, the Export button (Markdown) and the JSON button next to it save your whole coaching portfolio through the native save dialog — every artifact and its version history, Core Profile, bookmarks, memories, goals and check-ins, session feedback and insights. Nothing is cut off: every feedback entry, bookmark and artifact goes into the file (the safety ceiling is 100,000 of each, far above any real history). If something has to be left out — for example your insights when no Anthropic API key is set up — the app shows a warning ("…This is not a complete backup.") instead of the green "Your AI Coaching, Journal and Life Inventory data was exported." message, a Markdown file opens with a note naming what is missing and why, and a JSON file lists it under
omitted. An emptyomittedmeans the export left nothing out that it knew about; it does not cover an export of a single artifact or more than 100,000 items of one kind. A file made before this change has noomittedat all. Inomitted,"insights": "no-api-key"means insights are stored but no API key is set up to read them, and"excluded-by-choice"marks a section you switched off. Insights belong to the Omniscio account that owns your coaching data, not to the API key: with a key set up the file carries every insight you have, including ones saved under an earlier key, and a different Omniscio account signed in on this computer gets none of them. The export no longer writes"insights": "different-account"; a file made before this change may still carry it. The CLI export (POST /ai-coaching/export) returns the sameomittedmap next to the file content. An agent's CLI export asks you first: it raises an approval card in your inbox and gets no data until you approve, and then only one copy of exactly what it asked for. The desktop Export and JSON files also hold your whole Journal (every entry with its tags and guided conversation, including entries still in the trash) every Life Inventory take with its ratings, every custom interview topic you created (its title, description and date, under Custom interviews;customInterviewsin JSON), and every coaching conversation you can still open (live or archived, as the same transcript a single-session export writes, under Coaching conversations;coachingConversationsin JSON), because Delete all data removes those too and its Export a backup first button runs this same export. If the conversations would push the file past the 100 MB limit the export fails with a message saying so, rather than leaving any out. An export of a single artifact and the CLI export stay coaching-only. A Markdown file shows each older artifact version, journal entry and Life Inventory take under your local date; a JSON file keeps the exact timestamp. - Re-open old work. The Archived section lists every session that's been archived (whether from this sub-sidebar or from elsewhere in the app — archive is a session-level state, not a sidebar concept). Click any row to re-open the conversation read-only.
Personal & Professional tracks
When your interview library spans both personal and professional territory, the Dashboard's browse area groups the interviews into two labeled sections — Personal (relationships, identity, patterns, wellbeing) and Professional (career, ambitions, work) — with the usual theme groups nested inside each. An interview that fits both (foundations, ongoing check-ins, life-spanning goals) appears in each section, so nothing is hidden from either side.
- A small All · Personal · Professional selector sits above the theme filter (each with a live count). Pick one to focus the Dashboard on just that track; All (the default) shows both sections. Your choice is remembered on that device.
- The first time a real split is available, a gentle one-time card asks what you're here for — Personal, Professional, or show me everything — and sets the selector for you. Dismiss it any time; it won't nag again.
- If none of your interviews are specific to one track (for example a general-purpose library), the Dashboard simply keeps the flat theme view — no track sections, no selector — so nothing changes for you.
This is a way to stay in the zone you're comfortable with: someone who only wants career coaching can filter to Professional and never see the more inward interviews, while someone doing deeper personal work sees everything relevant to them. It's a view preference only — no interview is ever removed, and switching back to All brings everything into view.
How it behaves
Daily welcome message
The coaching Dashboard opens with a personalized greeting — "Good morning, {first name}" (or "Welcome back" if no display name is set) — plus a daily rotating message drawn from a pre-generated pool. The message changes once per day (deterministic by date, not on refresh) and stays the same all day.
A Daily message style dropdown next to the greeting lets you pick the category: Motivational Quotes (default), Coaching Insights, Mindfulness Prompts, Stoic Philosophy, Humor & Light, Gratitude Prompts, or Challenge of the Day. Your choice persists as the coachingDailyMessageStyle setting. Each category has 30 messages; the daily pick is a deterministic hash of the date + category, so you see the same message all day and a new one tomorrow.
The banner sits above the progress header in DashboardHeader.tsx and renders responsively (stacked on mobile, side-by-side on desktop). The message pool and utilities live in daily-welcome-messages.ts; the component is DailyWelcomeBanner.tsx.
Interview depth adaptation
The coaching system automatically adapts how deeply it probes based on the user's engagement history. Three signals are measured: the average length of your own replies in the coaching sessions you have answered (the interview's opening prompt, AI pause summaries and machine-sent messages are not counted), the number of coaching sessions you have actually answered, and redo count (topics answered in more than one session). These produce a depth level — surface, moderate, or deep — which is injected as a <depth_adaptation> section in the coaching CLAUDE.md alongside the tone section.
- Surface: simple follow-ups, structured options, comfort-first. Typical for new users or those who give brief answers.
- Moderate: balanced open-ended and concrete follow-ups, matching the user's energy.
- Deep: layered follow-ups, respectful assumption-challenging, cross-session pattern references. For engaged users with history.
This is fully automatic — no user setting. The depth assessment is computed fresh each time a coaching session starts or its CLAUDE.md is refreshed. If the engagement signal query fails (e.g. empty DB), the depth section is silently omitted and the session proceeds normally. Internally, computeDepthAssessment() in ai-coaching-depth-tracker.ts is a pure scoring function, and getEngagementSignals() in coaching-analytics.ts runs the DB queries. A session counts as answered once the app has run a turn in it or you have written in it, so opening interview cards without replying leaves a new person on the new-user guidance.
What the coach is told about neurodivergence
No built-in interview tells the coach to look for signs that you are neurodivergent. Where an interview has guidance on adapting to how your mind works (for example ADHD or autism), the coach is instructed to use it only when you have told it, or asked it to, and never to look for signs of it, guess at it, label it or record it itself. The Parenting interview applies the same rule to a child: it brings up your child's possible neurodivergence only once you have said your child may be neurodivergent, or asked it to adapt to how your child's mind works, and it never looks for signs of it in your child. How a wording change like this reaches an install that already has an interview is described on the Daily Journal Check-In page.
What a Marriage or Parenting write-up records about other people
The Marriage and Relationship Deep Analysis interview, its Quick version and the Parenting and Family Dynamics interview write about people who never chose to be in your profile. So the Marriage write-ups call your partner "your partner" and never write their name, and never record a guess at your partner's sexual satisfaction. The Parenting interview no longer asks for your children's names, and its write-up refers to each child by order and age (for example "Older child, 9") and never writes a child's name or initial, and calls a co-parent "your partner" or "the other parent" without writing their name. An install that already has these interviews, including a copy stored under another id, gets the new wording on its next start.
API key requirement (and the missing-key banner)
The interview chat runs on whatever Claude account you're signed in with (an OAuth login is fine). But three background features — compressing each saved artifact into its Key Points / Summary / Micro levels, the per-artifact summary, and the Core Profile synthesis — call the Anthropic Messages API, which only works with an API key (an OAuth login can't use that endpoint). So those features need a working Anthropic API key configured in Settings → Accounts; the rest of AI Coaching does not.
How the app handles a missing key:
- You can turn AI Coaching on without a key. Settings → AI Coach → "Enable AI Coaching" (and any other switch that writes
aiCoachingEnabled) works whether or not a key is set, because the interviews run on your Claude sign-in. Interview and check-in Start buttons on the dashboard work without a key too; a check-in uses its ready-made opening prompt when there is no key to personalise it. Turning the feature off is always allowed. - Saving a key updates the banner immediately. A key saved into the isolated "use my own API key" slot is a settings write, not an account row, so it has no account change to announce; the settings pipeline announces the credential change itself (
src/main/services/settings-side-effects/credential-source.ts), which re-runs the availability probe. Saving, re-entering or clearing the key therefore moves the banner at once — including a re-entry over an already-stored key, which the renderer only ever sees as a redacted placeholder. - A continuous banner inside AI Coaching when no usable key is set. If the feature is on but no working key is configured, a non-dismissible amber banner sits at the top of the AI Coaching sidebar (and the mobile dashboard): "AI Coaching can't build your profile." Under that title it says that coaching chats still work, and names what pauses until a working key is added: the Key Points / Summary / Micro compression, Core Profile updates, and new memories, goals, insights and check-in reminders. A second line says these background summaries use your key, are billed per use, and are capped at about $0.70 a day (the four daily AI Coaching cost pools — 35 + 15 + 10 + 10 cents — in
src/main/services/ai/ai-coaching/cost/cost-cap.ts; if a pool changes, change that line's copy and its translations too). The dashboard's insights refresh button stays switched off until a working key is added. "Usable" means the key actually resolves and isn't blank — an account row whose stored key went unreadable (e.g. after a hardware/Windows change that reset the OS keyring) counts as missing. The banner clears itself the instant you add a working key in Settings → Accounts (no reload needed). - One card when the key's Anthropic account is out of credit. When Anthropic answers that the credit balance is too low, AI Coaching stops asking for that write-up's other summaries, and the startup catch-up stops at that write-up instead of trying every one — so a dry account costs one refused call per app start, not three per write-up. You get one inbox card per app run, "AI Coaching's background features need Anthropic credit", with an Open Accounts settings button. Your coaching conversations are not affected. After you add credit to your Anthropic account, the summaries start again the next time the app starts.
If you add a key after an interview already saved its artifact, that existing artifact won't retroactively compress — the levels generate on the next save, so re-save it (or redo the interview once) to fill them in.
Sessions section subgroups
The Sessions section sub-divides your in-flight interviews into status buckets — same precedence rules as the main projects sidebar so the UI behaves consistently across the app. From top to bottom, with later groups winning ties:
- Needs You — sessions where Claude is waiting on you (any status in the
ATTENTION_STATUSESset:needs_you,error,stalled). - Active — sessions currently
running,starting, orready(deferred-spawn sessions waiting for the user's first reply). - Snoozed — sessions you've snoozed until a future time.
- Scheduled — sessions with a scheduled response queued (Send Later).
- Paused — paused sessions.
Precedence is Snoozed > Scheduled > Needs You > Paused > Active — a snoozed session that also needs you appears under Snoozed, not Needs You, mirroring the main sidebar exactly. Empty groups are hidden. Archived sessions never appear here — they have their own section. Sessions in transient states like ended are not surfaced (consistent with the main sidebar's default view). ready sessions (deferred-spawn coaching interviews awaiting the user's first reply) ARE surfaced under the Active group.
If your sessions can't be read — say, right after you switch accounts and the connection drops — the Sessions and Archived lists show "This section failed to load." with a Retry button instead of a loading placeholder that never goes away. Retry reads both lists again, and the rows appear as soon as the read succeeds.
If one section of the coaching sidebar or dashboard hits a display error, only that section is replaced by the same "This section failed to load." bar with Retry, and the rest stays usable. The bar also prints a short Reference ID you can select and copy; quote it if you report the problem, because it ties what you saw to the crash record (see Logs and debugging). Retry clears it.
Semantic search
The Artifacts section of the sub-sidebar has its own search box above the artifact list, which looks across artifacts and bookmarks by meaning in the same way and lists the matches right under it. If a search cannot run, the box shows the error (usually "Search failed — please try again.") with a Try again button that runs the same search again, instead of saying "No matching artifacts found"; that message appears only when the search ran and found nothing. Your private Daily Check-in and Weekly Review entries never appear in coaching search results.
The Search button in the sub-sidebar opens a pane that lets you search across artifacts and bookmarks by meaning — powered by the same on-device ONNX embedding model used by global session search (MiniLM, cosine similarity, threshold 0.35). Type a question or phrase (minimum 3 characters, 300ms debounce), and results are ranked by semantic relevance with a percentage score badge. Click an artifact result to open it in the detail pane, or a bookmark result to jump to the bookmarks section. When the search box is empty, suggestion chips ("What are my core values?", "Show my growth areas", etc.) offer quick starting points. All processing happens on-device — nothing leaves the computer. The component is lazy-loaded (CoachingSearchPane.tsx) and backed by the AI_COACHING_SEARCH_ARTIFACTS IPC channel, which delegates to searchAllBySemantic() in the embedding service.
Any artifact result opens, however many artifacts you have, on a computer or a phone: one that is not among the 500 the sidebar loads is read on its own when you click it. If the artifact has been deleted since the results were listed, the pane says This write-up couldn't be opened. It may have been deleted. with a Back to write-ups button, instead of loading forever.
If the on-device model has not loaded yet, or cannot read your query, the pane says Search failed — please try again. with a Try again button rather than "No matches found", and the app starts loading the model, so a retry a few seconds later usually works. "No matches found" only ever means the search ran and nothing was close enough.
What persists between visits
The sub-sidebar's UI state is saved to localStorage under the key amc.aiCoaching.sidebar.v1; a separate key amc.aiCoaching.onboarding.v1 remembers whether you've dismissed the first-run welcome and which milestones you've already been congratulated for (so nothing re-celebrates on a return visit). The one exception is a write-up you delete: deleting it means that interview is no longer finished, so the next time you open the dashboard it forgets that congratulation, and finishing the interview again shows You finished "…": your summary is saved. once more. Restoring the write-up from Recently deleted can congratulate you again too. The welcome ALSO hides automatically once you've engaged — a completed or in-progress interview, or a Core Profile — independent of this per-device flag, so it never re-greets you on a fresh browser or phone. Pressing Start your first conversation does not set the flag — the welcome goes away because the interview you just started counts as engaging, so if the start fails it stays put and you can try again: the button greys out while the start is under way and comes back a few seconds later. Only Maybe later or the close button dismisses it. Specifically:
activeSection— which of the four sections is currently selected.expanded— whether each of Sessions / Artifacts / Archived is expanded or collapsed.
These survive switching to a different project and coming back, and they survive a full app relaunch. Selected session and selected artifact are NOT persisted — switching away and back will keep your section choice and expand/collapse state, but you'll need to re-pick the row. This is intentional: a stale "selected session" id from days ago would point at a session you're no longer thinking about.
Who owns your coaching data on this computer
Your AI Coaching, Life Inventory and Journal data is stored locally, and only the account that is signed in can use it. When a different Omniscio account signs in on the same computer, your data is moved to a set-aside area on this computer, and that person gets an empty coaching space of their own. When you sign in again, your data comes back exactly as you left it: write-ups, Core Profile, bookmarks, memories, goals and check-ins, Journal entries, Life Inventory takes and your custom interviews. Your coaching chats are hidden from the other account too. That includes the local control server's session list, search results and session digest, so an agent working for the other account is not handed your coaching chat titles, excerpts or summaries. They do not count toward that person's Sessions total or weekly chart, or toward how deep their coach goes. A coach that was still running for you cannot save anything into their space.
Things worth knowing:
Nothing is ever deleted by this. Moving data aside and back never removes a row. If something cannot go back (for example its coaching chat was deleted while you were away), it stays set aside rather than being thrown away.
The coaching folder on disk goes with the account that wrote it. The files your coaching space keeps on the computer — the instructions each coaching chat reads, the pull-menu versions of your write-ups, your Life Inventory detail file and the write-up files older versions saved — are a copy built from your coaching data, not the data itself. When a different account signs in they are removed, so nothing of yours is left where that person or their coach could read it. They are rebuilt from your own data the next time you sign in.
The previous account's coach stops when its chat is set aside. If it was still running, it is stopped, and no message reaches that chat until its own account signs back in: a reply is refused with the sign-in sentence, a scheduled message waits, and a resume after a usage limit waits too. Its "A coaching chat can't start" inbox card, which names the chat, is cleared at the same moment; it comes back only if the start still fails after its own account signs back in.
Signing out and back in as yourself moves nothing. The move only happens when a different account signs in. It runs in the background, so the window never freezes; on a large Journal the AI Coaching, Life Inventory and Journal screens show their usual loading state for a few seconds while it finishes.
While nobody is signed in, none of it can be read, by you, by the coach or by an agent. No message reaches a coaching chat either: a scheduled message or a resume after a usage limit waits until you sign back in.
The other account's coaching chats never show on your badge or in your Inbox. They leave every list the moment the account changes — signing out and a switch that could not finish included, not only a switch that completed — even when one of those chats was the one open on screen at the moment it happened.
They cannot be opened by naming them either. Asking for one by its session id — to read the transcript, export it, copy the chat, brief it, screenshot it, share it or send it to a cloud box — is refused, and an inbox card that came from one no longer carries that chat's id. The chat itself keeps full access to itself, and your own coaching chats are never affected.
Another account's coaching chat cannot be continued. If you open one, for example from session search, and reply, the coach does not start: the chat says "This chat is set aside on this computer. It belongs to a different account, or it is waiting to be claimed in AI Coaching." and it only picks up again when its own account signs back in — or, if the data was never linked to an account, when you claim it from the Unclaimed coaching data card. It never says to sign in: you already are. A coach that starts after a move always reads the signed-in account's space, never instructions left behind by the previous account.
Data with no recorded owner waits for you to claim it. On a computer where coaching data was never linked to an account, or where two accounts wrote into the same data, it is set aside when someone signs in, and an Unclaimed coaching data card appears at the top of the AI Coaching dashboard and the Journal. Only claim it if it is yours: Claim it as mine adds it to your account. No agent can claim it for you.
Your backups and data exports only ever hold your own data, and Delete all data only removes yours; another account's set-aside data is never touched.
Coaching chats stay between you and the coach
Every coaching chat you start — an interview, a redo, a Quick Check-in or a free-form chat — starts with Plain Speak cards and spoken narration switched off, so no short summary of what you told the coach is stored beside each reply. It is also closed to messages from other agents: in the session menu, Block agent messages shows a check mark. Your own replies still go through as normal, and you can turn the block off from that same menu item. Chats started before this change keep their old settings.
What happens around an interview — new built-in interviews after an update, how interviews relate to regular sessions, the wrap-up when one finishes, Quick check-ins, opening messages, Bookmarks and Goals — is on part 5 of this page.
Related
- cli-ai-coaching.md — manage coaching artifacts and start interviews from an AI agent over the CLI control server.
- projects-sidebar.md — how virtual projects (including AI Coaching) sit in the sidebar and grouping behavior.
- snooze-a-session.md — Snoozed bucket above; same snooze mechanism as anywhere else in the app.
- send-later.md — Scheduled bucket above; powered by the same Send Later queue.
- bulk-select-sidebar.md — Shift+J/K bulk select works on AI Coaching session rows the same way it does in the main projects sidebar.
This page is split across four more parts: part 2 covers the artifacts an interview produces and everything you can do with them, part 3 covers the memory, Core Profile and analytics built from them, part 4 covers the CLI control-server surface and the implementation pointers for agents, and part 5 covers what happens around an interview — the wrap-up, Quick check-ins, opening messages, Bookmarks and Goals.
Last verified 2026-11-02