v0.4.1

MemoryMap AI 0.4.1

Works without AI. Ask and Chat answer from your own notes when no model is running. Every sentence is quoted from a note and cited, and the AI still answers whenever a model is available.

New

• A composer that talks like a person: reads casual questions, typos and slang, answers several questions at once, replies naturally to "ok", "thanks" or "lol", and asks "Did you mean…?" when unsure. Choose a natural or professional voice.
• Tidy: nine reviews that need no AI (weak links, stray tags, duplicates, empty notes and more), each undone in one step.
• Embedding models: pick from 15 in Settings, each with its licence. Switching runs in the background while search keeps working.
• Graph: Organic, Clusters and Galaxy layouts, clearer link lines, two-way links, Reshuffle and Unlink.
• Templates: 12 new board templates and 8 new mind map templates.
• Meetings: one note with Agenda, Notes, Decisions and Action items; record into it and turn action items into reminders.
• Password recovery: a recovery key resets your password and keeps your private notes.
• Background tasks: live progress, named steps, a log and an always-visible Stop on every job.

Improved

• With a model running, it reads only the relevant parts of your notes, so prompts are about 17% smaller.
• You can refuse or cancel the automatic install of search by meaning; a stopped install leaves nothing behind.
• Link reasons say what two notes have in common.
• One icon for AI and one for the agent; tooltips show keyboard shortcuts.
• Ctrl+K is for commands; content search is in Find anything.
• Redesigned OCR workspace, Settings sidebar, Dashboard, New board dialog and model cards.
• Atlas breathes, reacts more naturally to taps and drops in when you unlock.
• New project website.

Fixed

• The Guide answers again, with or without a model.
• Menus, dialogs and focus rings no longer get cut off or open off-screen.
• Items dropped onto boards land where you put them; Shift-select adds to a selection.
• Undo works on changes to several notes at once.
• Night shift's Run now reads as far as the scheduled run.
• The first note in a new notebook no longer gets stuck on "Filing…".
• Many layout fixes across Notes, Library, Chat, tags, the lock screen and the graph.
• Several slow text patterns are now instant.

v0.4.0

MemoryMap AI 0.4.0

A notebook that files itself, now checked end to end, with or without AI.

Highlights

• Capture and filing fixed and verified. Notes save, file, tag and stay findable after a reload. 49 browser tests now guard these flows on every change.
• Works with no AI model. Filing by your own notes' words, search, Ask with cited sources, and reminders from plain words.
• Honest filing certainty. Never shows 100%, and offers other categories when unsure.
• Nothing lost. Unsaved note and document edits survive a reload or closed window.
• Mind maps and boards. Proper hierarchy and looks by level, an emoji and icon library, and drag to delete with Undo.
• Polish. The Writing room, chat progress that says what Atlas is doing, new Atlas avatars, subtle animations with their own switch, and VS Code-style line copy in the editor.
• Security. HTTPS for other devices, stronger passwords, private notes kept out of answers and backups, and no network use before you allow update checks.
• Windows. Installer audited, and every build now installs, upgrades and uninstalls in CI.

Full details: CHANGELOG.md

v0.3.32

What's Changed

• ci: bump github/codeql-action from 4.38.1 to 4.38.2 by @dependabot[bot] in #161
• deps: update sqlalchemy requirement from <3.0,>=2.0.54 to >=2.1.0,<3.0 by @dependabot[bot] in #160
• deps: update uvicorn requirement from <1.0,>=0.53.0 to >=0.54.0,<1.0 by @dependabot[bot] in #159
• deps: update fsspec requirement from <=2026.7.0,>=2026.7.0 to <=2026.9.0,>=2026.9.0 by @dependabot[bot] in #158

Full Changelog: v0.3.31...v0.3.32

v0.3.31

Full Changelog: v0.3.3...v0.3.31

v0.3.3

What's Changed

• v0.3.3: Atlas and the corner companion, faces for everyone, optional sign-in, the app.js split, addressable views, and a release-wide polish pass by @Braydenh563 in #157

Full Changelog: v0.3.2...v0.3.3

v0.3.2

What's Changed

• Review the outside changes, restore the whiteboard, and a night of measured fixes by @Braydenh563 in #149
• deps: update uvicorn requirement from <1.0,>=0.52.4 to >=0.53.0,<1.0 by @dependabot[bot] in #151
• deps: update tzdata requirement from >=2026.3 to >=2026.4 by @dependabot[bot] in #152
• deps: update alembic requirement from <2.0,>=1.19.2 to >=1.20.0,<2.0 by @dependabot[bot] in #153
• deps: update sentence-transformers requirement from <7.0,>=6.0.1 to >=6.1.0,<7.0 by @dependabot[bot] in #154
• deps: update sqlalchemy requirement from <3.0,>=2.0.52 to >=2.0.54,<3.0 by @dependabot[bot] in #155
• ci: bump github/codeql-action from 4.38.0 to 4.38.1 by @dependabot[bot] in #156
• 0.3.2: the open ledger finished, a security pass, the map render pass, and a full disk that no longer loses notes by @Braydenh563 in #150

Full Changelog: v0.3.01...v0.3.2

v0.3.1

Full Changelog: v0.3.0...v0.3.01

What's Changed

• Potential fix for code scanning alert no. 406: Empty except by @Braydenh563 in #145
• Potential fix for code scanning alert no. 408: Explicit returns mixed with implicit (fall through) returns by @Braydenh563 in #146
• ci: bump github/codeql-action from 4.37.9 to 4.38.0 by @dependabot[bot] in #148
• deps: update alembic requirement from <2.0,>=1.19.1 to >=1.19.2,<2.0 by @dependabot[bot] in #147

Full Changelog: v0.3.0...v0.3.01

What's Changed

• Potential fix for code scanning alert no. 406: Empty except by @Braydenh563 in #145
• Potential fix for code scanning alert no. 408: Explicit returns mixed with implicit (fall through) returns by @Braydenh563 in #146
• ci: bump github/codeql-action from 4.37.9 to 4.38.0 by @dependabot[bot] in #148
• deps: update alembic requirement from <2.0,>=1.19.1 to >=1.19.2,<2.0 by @dependabot[bot] in #147

Full Changelog: v0.3.0...v0.3.01

What's Changed

• Potential fix for code scanning alert no. 406: Empty except by @Braydenh563 in #145
• Potential fix for code scanning alert no. 408: Explicit returns mixed with implicit (fall through) returns by @Braydenh563 in #146
• ci: bump github/codeql-action from 4.37.9 to 4.38.0 by @dependabot[bot] in #148
• deps: update alembic requirement from <2.0,>=1.19.1 to >=1.19.2,<2.0 by @dependabot[bot] in #147

Full Changelog: v0.3.0...v0.3.01

v0.3.0

What's Changed

• Release 0.3.0: one design system, Atlas, the audit, and the packaged app fixed by @Braydenh563 in #144

Full Changelog: v0.2.2...v0.3.0

v0.2.2

What's Changed

• Add §114 product strategy analysis to ANALYSIS.md by @Braydenh563 in #141
• deps: update sentence-transformers requirement from <7.0,>=6.0.0 to >=6.0.1,<7.0 by @dependabot[bot] in #143
• v0.2.2 — toolbar dropdowns, OCR workspace, background agent control, and a design-system pass by @Braydenh563 in #142

Full Changelog: v0.2.1...v0.2.2

v0.2.1

What's Changed

• v0.2.0 — OCR workspace, Obsidian vault import, live editors, and one kebab everywhere by @Braydenh563 in #140

Full Changelog: v0.1.7...v0.2.1