Item updates now log what actually changed (e.g. "status: open → done,
priority: low → high") in the activity metadata. The activity feed page
already renders meta.changes — this populates it with real diff data.
The activity list endpoint returned raw entries without item context,
causing the activity page to show bare "Created"/"Updated" verbs.
The dashboard already enriched entries via GetItem() lookups — now the
activity handler does the same, and the frontend reads top-level fields
with metadata fallback.
Cover all dashboard computation paths: empty workspace, summary counts,
active items sorting and capping, active phases with progress, overdue
items (due_date and end_date), blocked items via dependency links,
suggested next from active phases, isDoneStatus for all terminal states,
phase completion detection, orphaned tasks, recent activity, item refs,
multiple active phases, and 404 for nonexistent workspaces.
Webhooks:
- Full subsystem with HMAC-SHA256 signing, event filtering, auto-disable
after 10 failures, test delivery endpoint
- Migration 010_webhooks.sql, model, store CRUD, dispatcher with tests
- Wired into item create/update/delete/move and comment create handlers
Dashboard API:
- active_items: Returns actual in-progress items with refs and priorities
- active_item_count on collections (excludes terminal statuses)
- Blocked item detection in attention (via dependency links)
- isDoneStatus expanded to include cancelled/rejected/fixed/implemented
Saved Views:
- Store CRUD, API handlers, routes for per-collection saved views
- View config stores filters, sort, view_type
Activity:
- Source filtering support (web/cli/agent) in activity list endpoint
Full-stack feature: move items between collections (e.g., idea → task)
with automatic field migration.
Backend:
- Field migration engine (items/migrate.go) maps matching fields,
handles type conversions, drops incompatible fields, applies defaults
- Store method updates collection_id and assigns new item_number
- POST /api/v1/workspaces/{ws}/items/{slug}/move endpoint
- Activity logging with "moved" action and from/to metadata
- 6 migration unit tests covering type matching, conversion, and edge cases
CLI:
- pad move <slug> <target-collection> [--field key=value ...]
- Accepts singular collection names (task, idea, bug, etc.)
Web UI:
- "Move to..." dropdown on item detail page
- Shows all collections except current with icons
- Redirects to the item's new URL after move
Field migration rules:
- Same type: transfer directly (validate select options)
- Compatible types (text↔url, number→text, select→text): auto-convert
- Incompatible types: drop silently
- Missing required target fields: apply defaults or error
- Fix SQL injection in sort parameter: validate field names against
alphanumeric regex before interpolating into json_extract queries
- Restrict CORS to localhost origins only (remove http://* wildcard)
- Change Makefile HOST default from 0.0.0.0 to 127.0.0.1
- Add Apache-2.0 license field to web/package.json
Export/Import:
- `pad export -o file.json` exports workspace (collections, items,
comments, links, versions) to portable JSON
- `pad import file.json --name X` creates new workspace with
regenerated UUIDs and remapped relations
- GET /workspaces/{slug}/export and POST /workspaces/import endpoints
- "Download JSON" button in workspace settings
- Import tab with drag-and-drop in the create workspace modal
- Full transaction wrapping for atomic imports
Create Workspace Modal:
- Extracted from sidebar dropdown into a proper centered modal
(sidebar's CSS transform was trapping fixed-position elements)
- Rendered at root layout level via uiStore flag
- Create and Import tabs with template picker and file drop zone
- Print suggested /pad prompts after `pad init` creates a new workspace
- Add `pad onboard` command that detects project tooling (language, build system,
test runner, CI, linter) and suggests matching conventions from the library
- Replace empty workspace welcome box with OnboardingChecklist component showing
a 4-step guided setup with progress bar and /pad prompt hints
- Add contextual tips with /pad prompts to empty collection states
- Add onboarding workflow to /pad skill for agent-driven codebase analysis
- Fix relation fields storing slugs instead of UUIDs: server now resolves
slugs/refs to UUIDs for relation-type fields on both create and update
URLs now use collection prefix + item number (e.g. /ideas/IDEA-15) instead of
long slugs. Backend ResolveItem() accepts either format for backwards compat —
old slug-based URLs and bookmarks still work. Updated all frontend link
generation (item cards, sidebar, search, wiki-links) to use refs.
Redesigned the EditCollectionModal to support full schema evolution: editable
field labels, delete fields, reorder with up/down buttons, and per-option
chip editing for select/multi_select fields. When select options are renamed,
a bulk migration updates all existing items via SQLite json_set/json_extract.
Pad — project management for developers and AI agents.
Single Go binary with embedded SvelteKit web UI, SQLite storage,
CLI, and Claude Code /pad skill integration.
https://getpad.dev
xarmian