Changelog
Every release, newest first. These are the notes the app itself shows under Settings → About after an update, rendered from the same files.
v1.20.1 —
Features
- You can now copy or cut a branch and paste it into another map, with all internal references preserved and Undo restoring the original.
- Dragging a node marks it as your placement, and adding or extending nearby nodes no longer moves it; Undo restores the previous state.
- Deleting or moving a node now tidies the remaining siblings into a clean arrangement while keeping any manually placed branch in place.
- Chats that exceed the model's context window now send the most recent messages and note how many older ones were left out.
- Reopening a map restores the zoom level, position, and last viewed sheet you left it at.
- The library now lets you sort maps by last opened, last changed, or name, remembers your choice, and shows the relevant date on each card.
- Every edit, undo, and redo now marks the map as changed, and Ctrl+Z undoes while Ctrl+Shift+Z or Ctrl+Y redoes a map edit.
Fixes
- Log files no longer contain your node titles, prompts, replies, search terms, or other personal data.
- A stuck AI agent turn can no longer block all other requests by generating endlessly without a token limit.
- Deleting a node now removes its children, chat history, contributions, and edges, and Undo restores all of them exactly.
- Semantic search now ranks results relative to the best match, filtering out unrelated topics and weak collisions.
v1.20.0 —
Highlights
- Find arXiv papers for any node in your map, get AI answers grounded in them with cited sources, and keep followed topics up to date — in any language.
Features
- Every prompt now has a Resend button that lets you send it again to the chosen model, with an option to undo the replaced reply.
- Links in notes, details, and chat replies are now preserved and open when tapped instead of appearing as plain text.
- Collapsed nodes now show a badge with the number of hidden children; tapping it expands the node.
- Zoom controls, keyboard shortcuts, and a Fit to screen button are now available on the canvas.
- The About section now lists the other platforms the app is available on, with a link to the download page.
- A Features & setup guide is now available from the library menu, the empty library view, and Settings → About.
Fixes
- A settings read that landed mid-save could return an empty map and wipe all settings; saves now write to a temp file first so readers always see complete data.
- Local model replies no longer run for minutes without stopping; extend and research now have a reply length limit.
- The model-activity overlay no longer gets stuck on screen and blocks interaction after a run is abandoned.
- Switching back to a provider now restores the model you previously selected for it instead of falling back to a default.
- Importing Freeplane and XMind maps now preserves named styles, edge colours, compact layout, and note icon and tag display settings.
- Saving settings no longer crashes if the screen is closed while the save is still in progress.
v1.19.0 —
Features
- Use Claude from Anthropic or OpenAI as your AI provider with a curated model list, alongside Ollama and LM Studio. Everything local keeps working exactly as before, and the app asks before anything leaves your device.
v1.18.1 —
Features
- Three more local models to pick from — Muse Glimmer, Qwen 3.8 (27B) and, on LM Studio, Bonsai 27B.
- The app now tells you when a model is being loaded or freed —.
- Undo now takes back an applied 'Extend with AI' from the.
- Cutting a node now removes it from the map right away, with a banner to cancel the cut or paste it onto another node.
- A research note now shows which model wrote it, and keeps saying.
Fixes
- A model's deeply nested bullet list is now typeset in chat instead of appearing as a grey box of raw LaTeX.
- The 'keep or undo?' bar for AI changes no longer runs off the.
- A map-extend from bonsai-27b keeps the nested nodes it asked for instead of losing them to a stray quote.
- When a prompt needs more context than the graphics card can hold, the app now says the reply is being generated on the CPU instead of leaving you watching a blank screen.
- When a model's reply cannot be read as map changes, Extend.
- A large map no longer stops working with Ollama after the first slow reply — the app remembers what your card can hold on the GPU without refusing prompts it has already answered.
- The node style panel now opens centred on desktop instead of stuck to the bottom of the window.
- The AI settings no longer show a leftover 'Found N results'.
- Dialogs now show a divider under the title or above the.
- The interface text size you pick in Settings is kept between launches instead of resetting to normal.
- AI features no longer reload the model between them, so an AI.
- Formulas in AI answers and notes now come out typeset, not as raw LaTeX in a code box.
- AI notes now show formulas typeset, the way the chat does.
- The node style panel now has its Cancel and Apply buttons at the bottom, like every other dialog in the app.
v1.18.0 —
Features
- On Android, the model list for Ollama and LM Studio is the same as desktop, available before entering server address.
- On Android, connect to Ollama or LM Studio by entering just your computer's address — app fills in port and warns if not on local network.
- Nodes now stay lined up with siblings when renamed, restyled, or have notes/links/details/icons/tags added.
- A reasoning model asked to extend a node gets room to think before answering.
- Ask about large maps on local models and it sees the whole map — app gives Ollama/LM Studio needed context window.
- Esc now closes every dialog, including note and details.
- Added Qwen 3.6 (35B-A3B), a reasoning model that shows its thought process above chat replies.
- The user guide now shows how to open firewall port for sharing Ollama/LM Studio on Windows, Linux, macOS.
Fixes
- Changing the AI provider or model frees previous model from memory.
- When local models refuse requests or return nothing, chat and AI actions explain why.
- Check your AI connection from Settings before relying on it.
v1.17.23 —
Features
- Added Qwen 3.6 (35B-A3B), the highest-quality model on the list. It reasons before it answers — open the Reasoning line above a chat reply to see how it got there.
- The in-app user guide's feedback and licence links work again.
- Offer LM Studio as a selectable AI provider.
Fixes
- Stop models echoing <nodeId>.
- Unwrap bracketed ids, and stop LM Studio starving its own models.
- Append a new node below the siblings it joins.
- Use measured LM Studio download sizes.
- Offer a store install only what its store actually serves.
v1.17.22 —
Features
- feat(test-menu): run desktop and Android integration suites in parallel
- feat(canvas): move a node by cutting and pasting it onto another
- feat(layout): store node coordinates instead of re-deriving them on open
- feat(chat): inspect raw LLM prompt and reply per message
Fixes
- fix(ai): resolve mistral's undeclared "t1" parent instead of flattening
- fix(android): close DB connections between tests; unfocus before Quill dispose
- fix(layout): space nodes by their painted boxes, not an estimate
- fix(vm-test): exclude .snap-build from source rsync
- fix(snap): refresh the SSO discharge instead of demanding a new login
- fix(snap): export a year-long store credential instead of a one-day one
- fix(snap): explain Snap Store auth failures instead of dumping the JSON
- fix(snap): resolve the app icon from inside the snap
- fix(snap): register the desktop file so the launcher entry is exported
v1.17.21 —
Features
- feat(update): detect the install channel; never self-update a store build
- feat(security): encrypt stored credentials at rest behind one passphrase
- feat(prereq): check/install gcloud; harden the YouTube OAuth login
- feat(snap): add Snap Store install-metrics report
- feat(snap): auto-resolve latest beta revision when promoting to stable
- feat(snap): automate Snap Store listing + align all descriptions
- feat(snap): mirror the .deb launcher's GDK/GL handling via command-chain
Fixes
- fix(snap): force the exec bit on the command-chain wrapper
- fix(metrics): read snap install breakdowns on one shared bucket
- fix(snap): register name using the exported store credential
v1.17.18 —
Fixes
- fix(snap): read exported store credential through the container
- fix(clean): remove Python caches from artifact cleanup
- fix(snap): make the snapcraft manifest build-ready
v1.17.17 —
Features
- feat(screencast): record feature video for propagation
Fixes
- fix(linux): resolve launcher APP_ROOT at runtime so Flathub .deb runs standalone
v1.17.16 —
Features
- feat(screencast): record feature video for propagation
Fixes
- fix(linux): resolve launcher APP_ROOT at runtime so Flathub .deb runs standalone
v1.17.15 —
Fixes
- Maintenance and stability improvements.
v1.17.14 —
Features
- feat(macos): ship a universal (arm64 + x86_64) build
v1.17.13 —
Features
- feat(android): install built APK on a physical device from the menu
- feat(android): LAN Ollama connectivity, edge-to-edge toolbar, real-phone tests
v1.17.12 —
Features
- feat(android): install built APK on a physical device from the menu
- feat(android): LAN Ollama connectivity, edge-to-edge toolbar, real-phone tests
Other changes
- chore: test build v1.17.12.466
- chore: test build v1.17.12.465
- chore: test build v1.17.12.464
- chore: test build v1.17.12.463
- docs(site): note that local Ollama avoids cloud AI provider costs
- test(nightly): realign N-11/N-30/N-37 with recent feature changes
v1.17.11 —
Features
- feat(mobile): remote-only Ollama config + follow-device orientation
- feat(theme): add selectable light/dark/system theme with contrast-safe canvas
- feat(release): publish SHA-256 checksums.txt + on-site verify guide
- feat(canvas): add per-node "Realign children" context-menu action
- feat(ai): show model size in selector before download
- feat(dist): add Download stats menu/CLI pulling Cloudflare edge analytics
Fixes
- fix(ollama,linux): reliable zero-touch install + correct app icon
Other changes
- chore: test build v1.17.11.460
- chore: test build v1.17.11.459
- chore: test build v1.17.11.458
- test(ai): cover model-size feature
- chore: test build v1.17.11.457
- chore: test build v1.17.11.456
- chore: test build v1.17.11.455
- docs(site): explain how to open the app past macOS Gatekeeper on first launch
v1.17.10 —
Features
- feat(ai): warn when the Ollama endpoint leaves the local network
- feat(site): auto-publish the website with each release + screenshot gallery
Fixes
- fix(about): show What's New for the running version, not a stale feed release
- fix(publish): clear orphaned multipart uploads left by an interrupted publish
- fix(publish): stop guillotining the release upload at 30 seconds
Other changes
- docs(site): note the macOS build is Intel (runs on Apple Silicon via Rosetta 2)
v1.17.9 —
Fixes
- fix(release): tolerate the build-number bump a test build leaves behind
- fix(android): disable v4 signing to stop the stray .idsig in dist/
v1.17.8 —
Features
- feat(keystore): discover backups from the menu; install the tools from prerequisites
- feat(android): central signing + an encrypted keystore migration flow
Fixes
- fix: exclude the real keystore filename from git and the agent bundle
- fix(release): commit the whole version bump with the release tag
- fix(farm): never bundle raw key material to the build agents
- fix(release): block debug-signed APKs; stop building a pointless flathub .rpm
v1.17.7 —
Other changes
- chore: release v1.17.6
v1.17.6 —
Features
- feat(site): add a Preview mode so the site can be seen before the CDN exists
Fixes
- fix(release): refuse to publish a release built from stale or missing artifacts
v1.17.5 —
Features
- feat(about): ship the user guide and licence inside the app
- feat(export): hide export/share behind kExportEnabled until output improves
- feat(templates): four starter maps, authored in code
- feat(export): shareable map export — the growth loop
- feat(screenshots): capture phone-form-factor shots on the Android emulator
- feat(play): release signing, app bundle, and a measured size check
- feat(launch): feedback channel, store build flavour, AppStream metainfo, Flathub manifest
- feat(site): landing, privacy, license and comparison pages generated from identity
- feat(dist): single-source project identity; discovery-first channels
- feat(ai-chat): debug-channel toggle to gate raw LLM prompt/reply panels
- feat(ai-chat): style any chat message; consistent version title
- feat(editor): inline text styles for notes, details, and AI chat
- feat(dist): add test release channel, package manifests, and update UX
- feat(security): guard against plaintext writes on non-SQLCipher builds
- feat(build-farm): local MinIO build farm replacing GitHub CI
Fixes
- fix(macos): repair a truncated pub cache before building
- fix(macos): build Darwin plugins with CocoaPods only, not Swift Package Manager
- fix(macos): re-resolve CocoaPods so newly added plugin pods get installed
- fix(farm): run flutter pub get before build_runner on macOS/Windows agents
- fix(linux): bundle the real ONNX Runtime library, not a dangling symlink
- fix(test): scope each scenario's setUp hooks to its own group
- fix(library): move map actions to the app bar so they stop covering the list
- fix(flathub): make the Flatpak actually build, install and run
- fix(build-farm): harden agent + source bundling, restore version prompt
- fix(version): display the app version consistently across platforms
Other changes
- chore(farm): show 80 log lines on a failed job, not 25
- chore(macos): dump the Pods tree when a macOS build fails
- chore(identity): use a project feedback address instead of a personal one
- docs: record the pre-launch gaps as their own stage
- test(integration): cover the new features (N-35, N-36, N-37)
- docs: wire the phone screenshots into everything that consumes them
- chore(prereqs): check and install the Flatpak toolchain
- test(ai-chat): update N-09d for visible message action buttons
- docs(dist): document channel setup sequence; fix menu paths for two-level menu
v1.15.0 —
Features
- feat(canvas): hover card to reveal a node's full details
- feat(ai): add qwen2.5:32b curated model with real-LLM coverage
- feat(security): enable encrypted DB on Windows; handle stale credentials
- feat(security): encrypt database and settings at rest
- feat(tooling): automated presentation screenshot capture
Fixes
- fix(security): handle orphaned encrypted DB; speed up crypto tests
- fix(security): prompt and reload when switching to another encrypted DB
- fix(security): make key rotation crash-safe and recoverable
Other changes
- docs: add features-at-a-glance section to README
v1.14.8 —
Features
- feat(ui): add copy-to-clipboard buttons to LLM prompt and reply sections
Fixes
- fix(ai): harden Gemma/Qwen/Mistral strategies for large real-world maps
v1.14.7 —
Fixes
- fix(layout): place new children outward from parent's actual position
v1.14.6 —
Features
- feat(ai): add gemma4:e4b to curated Ollama models
Other changes
- test(real-llm): cover every curated model × AI feature with apply-and-verify
v1.14.5 —
Fixes
- fix(tests): stabilize vm suite orchestration and real-llm isolation
v1.14.4 —
Features
- feat(ai): support remote Ollama endpoints
v1.14.3 —
Features
- feat(ai): Ollama-only release surface with auto-migration
- feat(ai): in-app Ollama auto-install + idempotent provisioner pipeline
v1.14.2 —
Fixes
- fix(linux-update): install rpm packages silently
v1.14.1 —
Features
- feat(build): add parallel Linux rpm packaging
Fixes
- fix(build): prevent root-owned Flutter artifacts in Docker
v1.13.0 —
Features
- feat(ai): curated Ollama models with per-model strategies + lazy download
- feat(testing): run Linux integration tests on remote SSH-accessible VM
v1.12.3 —
Fixes
- fix(ci): commit embedding model so CI and release bundles include it
v1.12.2 —
Fixes
- fix(release): bundle onnxruntime in Linux .deb and verify packages
v1.12.1 —
Fixes
- fix(build): clean before Linux build and guard against missing libapp.so
- fix(update): install Linux update silently via pkexec and relaunch
v1.11.1 —
Features
- feat(license): add Elastic License 2.0
- feat(input): Enter sends AI chat message; Ctrl+wheel zooms canvas
Fixes
- fix(research): preserve paragraph line breaks when displaying existing notes
Other changes
- chore: rename AI menu items and bump build number
- perf(tests): run integration tests via combined entry points to avoid per-test recompilation
v1.10.0 —
Features
- feat(icons): render priority icons as distinct circled numbers
- feat(library): create new map via named root dialog
Fixes
- fix(tests): fix dialog cancel type mismatch and test isolation gaps
- fix(storage): debug mode always uses local data/ dir, ignoring SharedPreferences
Other changes
- chore(deps): downgrade transitive dependencies to stable versions
- chore: ignore .claude/ directory
v1.9.9 —
Fixes
- fix(linux): render on KDE+NVIDIA via Wayland + xsettings schema shim
v1.9.8 —
Features
- feat(migration): switch to existing destination DB instead of blocking
Fixes
- fix(script): clean generated artifacts after release
Other changes
- build: update version metadata and linux docker context
v1.9.7 —
Fixes
- fix(script): fix Invoke-PushLatest on Linux ($env:TEMP is null)
- fix(linux): fix GTK GL crash and .deb runtime deps on Ubuntu 24.04
v1.9.6 —
Fixes
- fix(storage): honor data-dir override in debug and tests
v1.9.4 —
Features
- feat(editor): add color picker dialog and full test wiring
v1.9.3 —
Fixes
- fix(storage): prevent db landing in install dir on release builds
v1.9.2 —
Fixes
- fix(installer): preserve shared_preferences.json across updates
v1.9.1 —
Features
- feat(canvas): right-click context menu on nodes
- feat(db): persist map-level AI chat history across app restarts
- feat(ai): add map-level extend and chat features; group in Map AI actions menu
Fixes
- fix(ui): remove duplicate Save button from Settings AI tab
- fix(storage): debug builds always use local data/ dir, ignore SharedPreferences
v1.9.0 —
Features
- feat(migration): database migration dialog with Windows file-lock fix
v1.8.7 —
Fixes
- fix(math): normalize indented bullet math to prevent code block fallthrough
v1.8.6 —
Fixes
- fix(ci): harden macOS DMG packaging step
v1.8.5 —
Fixes
- fix(math): handle double-escaped LaTeX delimiters
v1.8.4 —
Fixes
- fix(math): normalize latex rendering and add regression tests
v1.8.3 —
Fixes
- fix(ui): render multiline \( ... \) formulas as block math
v1.8.2 —
Fixes
- fix(test): isolate UI text scale storage
v1.8.1 —
Fixes
- fix(settings): guard UI text scale load for CI test env
v1.8.0 —
Features
- feat(settings): add global UI text scale control
- feat(ui): render LaTeX formulas in chat bubbles and node notes
v1.7.1 —
Other changes
- test: CI-safe parser tests, restore local full examples for menu 6, and stabilize freeplane tests with fixtures
v1.6.1 —
Other changes
- chore(ci): harden PR workflow with per-file unit tests and retry logic
v1.6.0 —
Features
- feat(ai): enrich node context and expose sent prompts in node AI flows
- feat(chat): render markdown formatting in assistant messages
- feat(chat): implement node chat persistence, error handling, and model preflight validation
- feat(testing): add manual metadata edit coverage and CI parity
Fixes
- fix(richcontent): freeplane richcontent import fix, regression tests, menu 6/test-runner hardening, and version bump
- fix(tests): automate map cleanup and resolve integration test brittle assertions
v1.5.3 —
Features
- feat(canvas): move Add note with AI into AI actions popup menu
Other changes
- ci: retry flaky test steps in PR gate
v1.5.2 —
Fixes
- fix: handle null result from Where-Object in release notes generation
Other changes
- ci: run PR gate on release tags
v1.5.1 —
Other changes
- ci: align local gate with PR workflow and enforce run cleanup
- chore: remove unused generated screenshots
v1.5.0 —
Features
- feat(ai,map): fix AI action routing and add manual notes
- feat: add auto-align and stabilize semantic search
Fixes
- fix(map-view): prevent auto-align overlap and map N-16
- fix(map-view): keep xmind manual node moves after auto-align
Other changes
- test: expand nightly scenarios and suite wiring
- chore(git): ignore all log files
v1.4.6 —
Features
- feat(ai-research): require confirmation before replacing existing note
Fixes
- fix(ci): run PR gate on release tags
v1.4.5 —
Features
- feat(testing): expand integration coverage and test wiring
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(release): restore action-triggered release tagging
- fix(parser): prefer content.json over content.xml to fix Central Topic.xmind version compatibility
- fix(ci): resolve analyzer findings
- fix(macos): raise deployment target to 14.0
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore: release v1.4.4
- chore(publish): exclude workflows from snapshot branch
- test(parsers): add comprehensive example files import regression suite with strict logging
- chore: release v1.4.3
- chore(release): prune old workflow runs
- chore: release v1.4.2
- chore: release v1.4.1
- chore(release): auto-clean invalid and old releases
- chore: release v1.4.0
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.4.4 —
Features
- feat(testing): expand integration coverage and test wiring
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(parser): prefer content.json over content.xml to fix Central Topic.xmind version compatibility
- fix(ci): resolve analyzer findings
- fix(macos): raise deployment target to 14.0
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore(publish): exclude workflows from snapshot branch
- test(parsers): add comprehensive example files import regression suite with strict logging
- chore: release v1.4.3
- chore(release): prune old workflow runs
- chore: release v1.4.2
- chore: release v1.4.1
- chore(release): auto-clean invalid and old releases
- chore: release v1.4.0
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.4.3 —
Features
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(ci): resolve analyzer findings
- fix(macos): raise deployment target to 14.0
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore(release): prune old workflow runs
- chore: release v1.4.2
- chore: release v1.4.1
- chore(release): auto-clean invalid and old releases
- chore: release v1.4.0
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.4.2 —
Features
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(ci): resolve analyzer findings
- fix(macos): raise deployment target to 14.0
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore: release v1.4.1
- chore(release): auto-clean invalid and old releases
- chore: release v1.4.0
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.4.1 —
Features
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(macos): raise deployment target to 14.0
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore(release): auto-clean invalid and old releases
- chore: release v1.4.0
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.4.0 —
Features
- feat(search): run semantic lookup on Enter
- feat(semantic-search): add embedding search UX and tests
- feat(import): split XMind workbooks into separate maps per sheet
- feat(test): add parser unit tests to dev script; use root topic as sheet title
- feat(import): fix UTF-8 decoding for multilingual maps; add XMind JSON support
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix(map-view): make root node title fit
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore(script): improve clean flow and menu formatting
- docs(readme): fix inconsistencies with actual implementation
- chore: release v1.3.1
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.3.1 —
Features
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore: release v1.3.0
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.3.0 —
Features
- feat(update): add release-driven auto-update with What's New notes
- feat(scripts): add -Publish parameter to mindmapchat.ps1
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(android): G-05 import test + Android emulator gate + DB path fix
- feat: add integration test scaffold and test menu
- feat: install app from GitHub release tag
- feat: add automated GitHub release pipeline
- feat(scripts): add clean option to remove build artifacts
- feat(scripts): add icon normalization pipeline with rembg and corner-sample bg removal
- feat(script): installer build, install, auto-setup in mindmapchat.ps1
- feat(canvas): drag-to-reparent nodes
- feat(settings): provider dropdown with live model list
- feat(canvas): replace 'Research with AI' with 'Extend with AI' debug dialog
- feat(ai): rich insertion actions + AI attribution badges
- feat(library): add map deletion with confirmation dialog
- feat(ai-chat): node-focused AI chat with per-node conversation history
- feat(canvas): show DETAILS richcontent as subtitle on node widget
- feat(storage): move DB and settings to configurable local data dir
- feat: rich node attributes - link/tag/attr display, icon editing, properties dialog
- feat: wire Node AI Research Panel into canvas and node widget
- feat: Node AI Research Panel
- feat(canvas): hierarchical layout, node dragging, uniform edge style
- feat(canvas): AI change highlighting with Keep / Undo banner
- feat(ai): two-pass map edit detection
- feat: initial implementation of MindMap Chat
Fixes
- fix: use build_number.txt in CI and add latest GitHub release install
- fix: include required icon asset in release builds
- fix(import): generate fresh UUIDs on every import to prevent ID collision
- fix: AI chat apply fails with partial tree - add tempId forward references
- fix(ai): robust JSON extraction for local LLMs (3-strategy parser)
- fix(canvas): prevent map jump and pan lockup after node drag
- fix(ai): allow Ollama and keyless providers; fix settings button layout
Other changes
- chore: release v1.2.0
- test: finish implementation of N-02, N-03, add logging, wire main_test.dart
- test: guarantee config isolation with setUp/tearDown hooks
- test: implement detailed integration tests and fix command stack stability
- refactor: inline push-latest.ps1 into mindmapchat.ps1
- chore: release v1.1.1
- chore: release v1.1.0
- chore: exclude push-latest.ps1 from repo
- docs: update README to reflect current implementation
v1.2.0 —
Features
- feat(ui): add tooltips for node indicators, sheet tabs, API key toggle, and empty-canvas hint
- feat(scripts): add -Publish parameter to mindmapchat.ps1
Fixes
- fix: use build_number.txt in CI and add latest GitHub release install
- fix(android): guard local data path on mobile platforms to avoid /data write failures
Other changes
- chore(release): align workflow build numbering and release/install script behavior