The Rust YAML parser only accepted boolean values (true/false) for the
archived and trashed fields. When the vault writes Archived: Yes or
Trashed: Yes (YAML string, not boolean), serde silently returned None
and the note appeared as non-archived/non-trashed.
Add a custom deserializer that accepts both booleans and string
representations (Yes/yes/YES/true/1 → true, No/no/false/0 → false).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The vault-api proxy maps Tauri commands to HTTP endpoints when a vault
API server is running. Without this, reload_vault bypassed the proxy
and Playwright route interceptors couldn't catch vault reload calls.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reload Vault (Cmd+K) now calls the new `reload_vault` Tauri command which
deletes the cache file before scanning, guaranteeing a full filesystem
rescan. Previously it called `list_vault` which used incremental git-based
cache updates that could miss recent changes (e.g. trashing a note then
reloading showed stale data).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Use untyped function references to avoid conflicts with
posAtCoords overloaded type signatures.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Covers clicking at 150%, 80%, and double-click word selection at 125%.
Verifies cursor lands near the click point (within first 30 chars of line)
and that word selection produces a non-empty range.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
CSS zoom on document.documentElement causes a coordinate space mismatch
between mouse event clientX/Y (viewport space) and Range.getClientRects()
(CSS space), breaking CodeMirror's click-to-position mapping. The previous
requestMeasure() fix only recalibrated cached geometry, not this mismatch.
New approach: zoomCursorFix extension patches posAtCoords/posAndSideAtCoords
on the EditorView instance to use document.caretRangeFromPoint() — the
browser's native, zoom-aware API — with coord-adjustment fallback.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The AI chat panel (AiPanel → useAiAgent) was sending each message as a
standalone request with no prior context. Root cause: useAiAgent.sendMessage
called streamClaudeAgent with raw text, never embedding history.
- Add agentMessagesToChatHistory() to convert AiAgentMessage[] to ChatMessage[]
- Embed trimmed history in each agent request via formatMessageWithHistory
- Use messagesRef/statusRef to avoid stale closures in async callbacks
- Also fix useAIChat (dead code path) with same ref pattern
- Update mock layers to detect history presence for testability
- Add Playwright smoke tests verifying history accumulates and resets
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Settings command is disabled in mock environment (onOpenSettings not wired),
causing the 'typing filters the command list' test to always fail.
Reindex Vault is always enabled and already tested in indexing-reindex-status.spec.ts.
Remove the early return in update_same_commit that skipped filesystem
validation when git reported no changes. Add prune_stale_entries to
finalize_and_cache so every vault scan path validates entries exist on
disk and deduplicates by case-folded path. Prevents ghost notes after
deleting files outside the app (e.g., via Finder).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move updateEntry() calls after handleUpdateFrontmatter/handleDeleteProperty
in handleCustomizeType, handleRenameSection, and handleToggleTypeVisibility
so React state only updates after the disk write succeeds. This prevents
state-disk divergence when writes fail.
Expand ARCHITECTURE.md "Three representations, one authority" section with
ownership rules, invariants table, and recovery mechanisms. Add
reload_vault_entry to the commands table (62 total).
Add Playwright smoke test for Reload Vault in Cmd+K palette.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a "Reload Vault" command that forces a full rescan from filesystem,
bypassing cache. Available via Cmd+K and Vault menu. Wired through
useAppCommands → useCommandRegistry and useMenuEvents.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Re-reads a single .md file from disk and returns a fresh VaultEntry.
Used after failed optimistic updates to restore the true filesystem state.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Any frontmatter field whose value contains [[wikilinks]] now renders as a
relationship chip automatically. Fields with plain-text values always render
as editable properties, even if they were formerly hardcoded relationship keys.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Verifies that searching "Writing" in Quick Open shows the exact title
match first, followed by prefix matches. Also adds Refactoring test
entries to mock data and demo vault.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The Rust parser now treats `type:` as the canonical frontmatter field
for entity type, with `Is A:` and `is_a:` accepted as legacy aliases.
Previously it was the other way around, creating an asymmetric
read/write cycle since the frontend and all 8800+ vault notes already
use `type:`.
- Flip serde attribute: rename="type", alias="Is A", alias="is_a"
- Update theme defaults, getting-started vault, and type definitions
- Add round-trip tests for both type: and Is A: parsing
- Update mock data and TypeScript tests to use canonical form
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Cache files are now stored outside the vault directory at
~/.laputa/cache/<vault-hash>.json, preventing them from polluting
the user's git repo. Writes use atomic tmp+rename to avoid corruption.
Legacy .laputa-cache.json files are auto-migrated and cleaned up on
first run.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Verifies that searching "Writing" in Quick Open shows the exact title
match first, followed by prefix matches. Also adds Refactoring test
entries to mock data and demo vault.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Title exact match gets exclusive tier 0 — alias exact match is capped
at tier 1, so a note titled "Refactoring" always appears above notes
with "Refactoring" as an alias or prefix. The 5-tier ranking is:
0=title exact, 1=alias exact, 2=title prefix, 3=alias prefix, 4=fuzzy.
Also adds ranking to editor wikilink autocomplete (enrichSuggestionItems)
and trims whitespace in searchRank comparisons.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Covers type change → move toast confirmation and type selector visibility
in the properties panel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When the user changes a note's type via the Properties panel,
the note file is automatically moved to the corresponding type folder.
Shows a toast confirming the move. No move if already in correct folder.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a new Tauri command that moves a note file to the folder
corresponding to its new type when Is A is changed. Handles:
- folder creation, filename collision (-2 suffix), wikilink updates,
and no-op when already in the correct folder.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Laputa as proof-of-work for Refactoring's credibility:
- Building publicly validates the author's authority to write about
building software with AI (not theory — demonstrated practice)
- Open source makes the work visible: GitHub commits are public evidence
- Success converts to reputation/acquisition for Refactoring via
sponsorships, paid subs, and brand authority
- Strategy: build the tool you describe, make the work visible
Claude Code must also test with pnpm tauri dev (not just Playwright)
when the task touches: filesystem, AI context pipeline, MCP server,
git integration, or native Tauri commands.
Playwright tests mock-tauri handlers — they cannot catch bugs in the
real file read/write layer. Phase 1b closes this gap.
Lesson from ai-chat-empty-body: bug was in MCP server reading from disk,
invisible to Playwright. Phase 1b would have caught it in attempt 1.
Add searchRank/bestSearchRank utilities that compute a tier (0=exact,
1=prefix, 2=fuzzy-only). Both useNoteSearch and WikilinkChatInput now
sort by rank tier first, then by fuzzy score, ensuring notes with exact
title or alias matches always surface above partial/fuzzy matches.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Require Claude Code to write a new task-specific Playwright test for
every task (not just run existing smoke tests)
- Test must fail before fix and pass after — proves coverage
- Clarify that Phase 1 is Claude Code's quality gate, not Brian's
- Brian's Phase 2 is a reinforcement check; if he finds a bug that
Phase 1 should have caught, that is a Phase 1 failure
Lesson from ai-chat-empty-body: 5 QA cycles happened because Phase 1
never verified that the AI actually received note content end-to-end.
Strongest possible answer to 'why are you the right person to build this':
- Generalist CTO who can build end-to-end
- 300+ articles = battle-tested PKM system at scale
- Refactoring distribution (~200K subscribers) = built-in audience
- Not theorized — the method is proven by the output that exists
The capture→organize→express framework is output-agnostic:
- Writers: evergreen notes as building blocks for articles
- Builders: project knowledge graph and shipped work
- Operators: procedures and responsibility systems
What varies is the expression layer; the discipline is universal.
handleSelectNote and handleReplaceActiveTab now check the in-memory
allContent cache before issuing a Tauri IPC call. Cache hits open the
tab synchronously (zero latency). Cache misses fall back to the disk
read and populate allContent via onContentLoaded so subsequent opens
of the same note are instant.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Two additions from Luca's published essays on note-taking:
- 'Knowledge has a purpose' section: notes exist to get things done,
not for abstract future use. Without purpose, the system collapses.
- Evergreen notes concept: atomic, timeless, reusable units of thought.
The most valuable layer of a mature vault.
- Organize phase clarified: weekly cadence, deleting >50% of captures
is normal and healthy, not a failure.
Complete rewrite structured around three pillars:
1. The problem (architectural + methodological)
2. The method (ontology, capture/organize, convention over configuration)
3. The foundation (local files, Git, AI-native architecture)
Key improvement: the document now explains *why* tool and method together
is the differentiating insight — not just a list of features and principles.
Includes the three-stage product trajectory and updated design principles.
Current state section condensed; full roadmap moved to ROADMAP.md.
VISION.md:
- New section 'The two-phase knowledge workflow: capture and organize'
- Explains capture (fast, frictionless, everywhere) vs organize (deliberate,
periodic) as fundamentally different activities
- Defines Inbox: a smart filter showing notes with no outgoing relationships
- Inbox Zero as the goal; connecting a note removes it automatically
- Replaces 'All Notes' as the primary navigation section
ROADMAP.md:
- New strategic direction #4: Inbox and capture pipeline
- Covers inbox UI, capture integrations (Chrome ext, iPhone, Readwise, voice)
- Add 'Product trajectory' section: 3 stages from personal PKM → indie
knowledge workers → small teams, with rationale for why the same
foundational model (local files + Git) enables all three stages
- Note that the knowledge ontology (Projects/Responsibilities/Procedures/
Notes/People/Events) maps equally well to personal and organizational use
- Describe workspace feature as the seed for future team access control
- Update design principles: add convention over configuration, semantic
properties, filesystem as source of truth; expand AI-native principle
to include AI-readability via shared conventions
Remove hardcoded /Users/luca/Laputa/ paths from resolveNewNote,
resolveNewType, and resolveDailyNote. All three now accept a vaultPath
parameter and build paths relative to the active vault. Added vaultPath
to NoteActionsConfig so the hook passes it through to all callers.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- ARCHITECTURE.md: new Design Principles section covering filesystem as
source of truth, convention over configuration, no hardcoded exceptions,
AI-first knowledge graph, and three-representation model
- ABSTRACTIONS.md: design philosophy intro + semantic field names table
documenting all conventional frontmatter fields and their UI behavior
Convention over configuration principle explicitly noted as serving
AI-readability: shared conventions make vaults navigable by AI agents
without bespoke per-vault instructions.
contextRef (useRef) was initialized at mount time and synced via useEffect,
which runs after paint. If contextPrompt was empty at mount (tab content
not yet loaded), sendMessage could read stale/empty context during the
window between paint and effect. Removing the ref and reading contextPrompt
directly in the useCallback closure eliminates the race entirely.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When viewing a Type note (e.g. "Project"), the Properties panel now shows
an Instances section listing all notes of that type, sorted by modified_at
descending. Trashed instances are excluded, archived instances are dimmed.
Display capped at 50 with count badge for large collections.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The frontmatter field for entity type is now 'type:' (not 'Is A:').
The Rust parser accepts both via serde alias, but all documentation,
examples, and new code should use 'type:'.
Updated: ABSTRACTIONS.md, ARCHITECTURE.md, PROJECT-SPEC.md, GETTING-STARTED.md
Three root causes identified and fixed:
1. JS semantics bug: `activeNoteContent ?? allContent[path]` used nullish
coalescing (`??`) which does NOT fall through on empty string ''. When
handleEditorChange temporarily overwrites tab.content with frontmatter-
only content during async content swaps, activeNoteContent becomes ''
and the fallback to allContent never triggers. Fix: change `??` to `||`.
2. Defence-in-depth: when body is still empty after fallback (Tauri mode
where allContent is {}), but wordCount > 0, the body field now includes
an explicit get_note instruction instead of being empty. This is more
reliable than the preamble instruction that Claude may skip.
3. Rust strip_frontmatter used `rest.find("---")` which matches `---`
anywhere in text (including inside frontmatter values like `title:
foo---bar`). Fixed to use `rest.find("\n---")` for line-boundary
matching. This ensures accurate wordCount for the fallback heuristic.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
