- update --prune: detect skills deleted upstream and move them to trash
(batch + single-target paths), clean registry entries after prune
- update without --prune: warn about stale skills with suggestion
- check: report "stale" status for skills whose subdir no longer exists
in the remote repo (previously misreported as update_available)
- Refactor updateTrackedRepo, updateRegularSkill, updateTrackedRepoQuick,
updateSkillFromMeta to accept *updateContext instead of 7-10 individual
params; delete dead updateSkillOrRepo function
- Add integration tests for prune (standalone, nested, all-stale,
registry cleanup) and project-mode prune
- Update website docs for update --prune and check stale status
For testing on UTM Windows VMs. Scans common shared folder
locations and presents interactive picker when multiple found.
make build-windows SHARED=~/Downloads
mise run build:windows ~/Downloads
Homebrew path: replace direct stdout pipe with TreeSpinner + bytes.Buffer
so brew output no longer breaks the tree layout. Captured output is only
shown on failure for debugging.
GitHub release path: switch download/UI-asset spinners from StartSpinner
to StartTreeSpinner with correct isLast flags, add tree-line prefix to
confirmation prompt, use StepEnd for cancel.
Add TreeSpinner.Warn() and TreeLine() helper to internal/ui.
Add devcontainer test-brew-upgrade script to simulate Homebrew path.
- Summary line: CRITICAL/HIGH/MEDIUM/LOW/INFO colored by severity
- Risk label: colored by severity (e.g. red for CRITICAL)
- Finding groups: severity-colored prefix, locations on separate
indented gray line (maxLocs reduced from 5 to 3)
- Add visual gap between summary block and finding details
- Replace SkillBox with inline tree steps (Desc, License, Location)
for single-skill and tracked-repo install output
- Add custom application-level filter to dir picker and skill select
TUIs, replacing bubbles built-in filter for consistency with
list/search/log TUI pattern
- Add renderSkillMeta/renderTrackedRepoMeta/truncateDesc helpers
Move Header + StepStart before filesystem scan and metadata reads so
users see immediate feedback instead of a frozen terminal.
- check: Header printed before GetTrackedRepos/GetUpdatableSkills
- check filtered: Header printed before resolve phase
- Add spinners during blocking scan/resolve phases
- Unify tree structure: Repo/Skill use StepContinue (not StepStart)
- Add "(tracked)" label to tracked repo in update output
Enrich the single-skill install SkillBox with the SKILL.md frontmatter
description field, giving users useful context instead of repeating the
skill name. License line appended when present.
- Add Description field to SkillInfo struct
- Use ParseFrontmatterFields to read description + license in one pass
- Extract buildSkillBoxDesc helper for the two single-skill call sites
pterm's animation goroutine and UpdateText() race on the same terminal
line — the former appends a timer suffix "(Ns)" while the latter does
not, causing per-frame length differences that manifest as flicker.
Two fixes:
- Disable ShowTimer (our Success/Warn already print elapsed time)
- Pad normalizeGitProgressMessage output to 40 chars so short messages
overwrite residual characters from previous longer messages
- Extract renderTUIFilterBar() shared by list, log, and search TUIs
- Extract renderPageInfoFromPaginator() for shared page info rendering
- Replace built-in bubbles filter with custom textinput filter in log
and search TUIs (matches list_tui pattern: / enter, esc clear, enter lock)
- search_tui: filter mode suppresses action keys (space/a/enter/s) to
avoid accidental checkbox toggles while typing
- Align status line left (remove extra indentation) across all TUIs
- Add ←→ page hint to all TUI help lines
- Unify pterm spinner to braille dot pattern (⣾⣽⣻⢿⡿⣟⣯⣷) with cyan
style, matching bubbletea spinner.Dot used in list_tui
Replace bufio.Scanner with json.NewDecoder which has no fixed line
length limit. Fixes "token too long" crash when audit results produce
JSONL entries exceeding the 64KB scanner default.
- Move data loading inside TUI with bubbles/spinner to eliminate
blank screen after entering alt screen
- Thousand separators in status line (108,749 skills)
- Page info in status line (Page 2 of 4,729) replacing raw "1/4944"
- Fix "(showing 1,000)" appearing with empty filter text
- Parallel ReadMeta with bounded worker pool (64 workers)
- Parallel git dirty checks in uninstall (8 workers)
- Use DiscoverSourceSkillsLite for uninstall --all discovery
DiscoverSourceSkillsLite skips frontmatter parsing during directory
walk and collects tracked repos in a single pass — significantly
faster for large skill collections.
ParseFrontmatterFields extracts specific YAML fields from SKILL.md
without parsing the full document, used by the TUI detail panel.
v0.16.0 moved update rollback logic from update.go to
update_handlers.go and changed the error message format.
Update the mutation target file and perl regex pattern so
the signal test finds the expected 2 mutation points.
handleGitInstall has two single-skill paths (subdir and non-subdir)
that always printed "✓ Installed: <name>" even in dry-run mode.
Branch on opts.DryRun to show "[dry-run] Would install: <name>" instead,
matching the multi-skill path behavior.
Fixes failing CI tests: TestInstall_Auth_HTTPSPublicWithToken,
TestInstall_RemoteGitHubSubdir_DryRun.
update.md: add --audit-verbose, --project/-p, --global/-g
uninstall.md: add --project/-p, --global/-g
Both sets of flags exist in source (printUpdateHelp, printUninstallHelp)
but were missing from the docs options tables.
Add missing replit badge + id anchor in supported-targets.md.
Add project-mode integration tests for audit, check, diff, and log
commands to close gaps found in codebase audit.
- Add flag validation step and quality checklist to runbook generation
- Add retrospective step after test execution for continuous improvement
- Document --init interaction, registry.yaml assertion rules, common flag mistakes
- Distinguish runbook bugs from real bugs in FAIL analysis
- Fix stale config.yaml skill assertions in uninstall_all_glob runbook
- Add registry_yaml_split runbook (7-step E2E for the registry.yaml refactor)
Skills are now stored in registry.yaml, not config.yaml. Update 10
integration tests to read from the correct file:
- install_group_test.go: 3 tests read registry.yaml after install
- uninstall_test.go: 3 tests read registry.yaml after uninstall
- uninstall_project_test.go: 4 tests write skills to registry.yaml
(project uninstall doesn't trigger config migration) and verify
registry contents after uninstall
Introduce a separate Registry type to hold skill entries (installed/tracked
skills) independently from config.yaml. This keeps config.yaml focused on
user-managed settings while registry.yaml tracks skill metadata.
Registry reuses the existing SkillEntry struct and follows the same
patterns as Config (schema comment, yaml.Marshal, validation on load).
Replace custom useApi hook with @tanstack/react-query across all 14
pages and InstallForm. Add react-virtuoso VirtuosoGrid to SkillsPage
for handling 46K+ skills without freezing the browser.
Key changes:
- Add queryClient with shared defaults (staleTime, gcTime, retry)
- Add centralized queryKeys factory for prefix-based invalidation
- Wire QueryClientProvider + ReactQueryDevtools in App.tsx
- Migrate AppContext to useQuery (shares cache with DashboardPage)
- Migrate all pages: useApi → useQuery, handlers → invalidateQueries
- SkillsPage grid view: VirtuosoGrid with useWindowScroll, overscan,
ScrollSeekPlaceholder, and React.memo on card components
- Add resolve.dedupe for React to prevent duplicate instance in dev
- Add vendor-tanstack-query manual chunk for build splitting
- Delete unused useApi.ts hook
Cover: latest backup round-trip, specific timestamp selection,
target not in backup, dry-run with force, force overwrite on
non-empty target, and no-force failure on non-empty target.
Modal now calls /api/restore/validate on open to show backup size,
symlink status, and file conflicts before confirming. Button text
dynamically changes to "Restore (overwrite)" when conflicts exist.
Removes hardcoded force:true — only sends force when user acknowledges.
Returns validation result with conflict details (existing files),
backup size, and symlink state so the UI can show informed restore
confirmation instead of blindly using force:true.
Grow the detail panel from 12 to 16 reserved lines so more metadata is
visible. Description now word-wraps to terminal width (up to 3 lines
with trailing "..." when truncated) and is followed by a blank line for
visual separation. Also display the license field from SKILL.md
frontmatter when present.
Replace static PS1 prepend with a dynamic PROMPT_COMMAND hook to prevent
prompt duplication when ssenv-shortcuts.sh is sourced multiple times.
The hook also auto-clears the prefix when SSENV_ACTIVE is unset (reset/delete).
Add .claude and .codex to ensure_layout() so new environments include
these directories by default.
Remove all survey/v2 dependency (7 call sites) and fuzzy direct usage:
- search.go: survey.Input → bufio.NewReader for simple text prompts
- init.go: survey.Select → inline numbered prompt for sync mode
- init.go: survey.MultiSelect → checklist TUI for target/agent selection
- init_project.go: survey.MultiSelect → checklist TUI for project targets
- install_prompt.go: survey.Select → checklist TUI for orchestrator mode
- install_prompt.go: delete dead promptSearchSelect function
- install_prompt.go: fuzzy → strings.Contains for filterSkillsByName
Add generic checklist_tui.go supporting both multi-select and single-select,
reusing the existing bubbletea/bubbles styling. Sync mode uses a simple
inline prompt (no alt-screen) for smoother init flow, with a copy-mode
hint when non-copy is selected. TUI selections now print confirmation
after exit so users see what they chose.
Single-skill check now shows the same tree-structured header (Source,
Items, Skill name, Source URL) and uses a spinner + StepResult instead
of a progress bar, consistent with how `update` renders single targets.
The // subdir syntax (e.g., repo//skills/foo) stores meta.Subdir with
a leading "/". git rev-parse and ls-tree paths never have leading
slashes, causing tree hash lookups to fail silently.
Strip leading "/" in getSubdirTreeHash(), resolveSkillStatuses(), and
handler_check.go. Add LeadingSlash unit test and E2E runbook.
Progress bar total was `len(repoInputs) + len(urlInputs)` which showed
"2/2" for 46K skills from 2 URLs — meaningless to users. Now uses actual
skill count so the bar shows "46321/46321" instead.
Previously `skillshare check` compared repo-level commit hashes, causing
all skills from a registry repo to show "update_available" whenever any
commit was pushed — even if the individual skill's files were unchanged.
Now each skill records its git tree hash (subdir-level SHA) at install
time. During check, a three-tier comparison is used:
1. Fast path: commit hash matches → up_to_date (zero extra cost)
2. Tree hash path: blobless fetch (~150KB) + batch ls-tree → per-skill
comparison; unchanged subdirs stay up_to_date
3. Fallback: no tree hash (old installs) → commit comparison (unchanged)
Also fixes pterm progress bar printing twice when it auto-stops at 100%.
Replace the single-select survey.Select prompt in `search` with a
bubbletea-based multi-select TUI, enabling batch installation of
search results. The new TUI supports space/a toggle, `s` for search
again, `/` filter, and esc to cancel — consistent with the existing
skill select TUI pattern from install.
Add batchInstallFromSearch() for sequential multi-skill installation
with per-skill error reporting.
- Replace promptMultiSelect's survey.MultiSelect with a full bubbletea
TUI (space toggle, a select-all, enter confirm, / filter, esc cancel)
- Unify TUI styling across all screens: brand yellow (#D4D93C) for
active item border+text, bright white for normal items
- Fix infinite loop when selecting (root) group in dir picker
- Compact single-line items with SetHeight(1) + SetSpacing(0)
- Pre-compute location strings and maintain selCount counter for
better toggle performance with large skill sets
- Remove inline ANSI from Title() to prevent filter highlighting corruption
- List TUI: single-line items with [tracked]/[local] badges, no description
Replace redundant "Updating N skill(s)" header + separator with the
same tree layout used by check: Header → StepStart("Source") →
StepContinue("Items") with repo/skill counts. Removes duplicate
"Updating" label that overlapped with the progress bar title.
groupSkillsByDirectory() used filepath.Dir() which always returned the
same parent regardless of recursion depth. Add a prefix parameter so
each level strips already-navigated directories before grouping.
Also fix stale assertion in TestCheck_RegularSkill_ShowsMeta to match
the summarized local-skill output introduced in 37f3425.