v0.17.5 blocked sync when target directories didn't exist, which
prevented first-time users on fresh CLI installs from syncing.
Sync now auto-creates missing directories and reports what it did
in CLI output, dry-run preview, and the Web UI. Init also creates
target directories when it detects an installed CLI tool.
Web UI: Sync Preview modal stays open after sync with a success
banner instead of immediately closing, so users can review results.
Closes#87
Covers: API routing with prefix, 404 without prefix, bare path redirect,
__BASE_PATH__ injection, SPA fallback, -b flag, env var fallback.
Also fix unused import in SyncPage.tsx (broke tsc).
Ref: #85
copyDirectoryWithState did not skip .git/ dirs, causing collect to fail
on git-cloned repos (e.g. obra/superpowers) — filepath.Walk would abort
on .git pack files, leaving only empty directories in source.
Add copyDirectorySkipGit() variant and use it in PullSkill. Existing
copyDirectory (used by copy-mode sync) is unchanged.
Use mdproof v0.0.5 per-runbook isolation to give each runbook its own
HOME and TMPDIR, eliminating inter-runbook state leakage that caused
4 of 22 runbooks to fail when run sequentially.
Key changes:
- Add ensure-skillshare-linux-binary.sh to build step (creates ss symlink)
- Use direct binary path in setup/teardown (bypass wrapper overhead)
- Create target dirs (~/.claude, ~/.codex) so --all-targets detects them
- Clean shared /tmp/ paths between runbooks
- Add references/extras.md covering all 4 subcommands (init, list,
remove, collect) plus sync extras and diff --extras
- Update SKILL.md with dedicated Extras recipes section and reference link
- Update status.md with diff --extras/--all flags
- Fix sync.md: extras supports project mode (not global only), correct
source path to include extras/ subdirectory
- Fix diff --extras not working in project mode (diff_project.go was
missing collectExtrasDiff integration)
- Replace DefaultDelegate with custom diffItemDelegate so Extras
separator renders as non-selectable group header (matching list TUI)
- Add skipDiffSeparator to prevent cursor landing on separator items
- Add Next Steps hints to extras detail panel in diff TUI
- Group Backup/Trash under "Storage" header in doctor output
- Remove extra leading spaces from doctor extras status lines
Extras commands had multiple issues with tilde (~) path handling — Go
doesn't expand ~ (that's shell behavior), so paths like ~/.claude/rules
were treated as literal relative paths, creating bogus ~/ directories.
Key fixes:
- Export config.ExpandPath() and call it before os.Stat / sync operations
in extras_list, sync_extras, extras_collect, and doctor
- Unify extras output headers to match target list style (single "Extras"
header instead of per-extra headers, fix double-icon bug in list)
- Add per-target reachability warning in doctor for unreachable extras
targets with expanded path display
Update all 3 E2E runbooks to use extras/<name>/ path layout, add /tmp
cleanup for idempotent re-runs, add migration test step, and create new
comprehensive extras_commands_runbook (18 steps covering init/list/remove/
collect/doctor/tilde/project-mode).
Improve cli-e2e-test skill with 4 new checklist items and 2 rules learned
from test friction (/tmp isolation, symlink write-through, cat>>
idempotency, extras path layout).
Repos like pbakaus/impeccable commit .claude/skills/ alongside their
source skills, causing discoverSkills to find every skill twice (36
instead of 18). Fix by deriving known target dotdir names (.claude,
.cursor, .skillshare, etc.) from targets.yaml project_path values and
skipping them during filepath.Walk.
Non-target hidden dirs (.curated, .system — used by openai/skills) are
still discovered as before. Injection pattern avoids circular import
between config and install packages.
- CHANGELOG/website: add gitlab_hosts config and SKILLSHARE_GITLAB_HOSTS
env var to New Features section, add HTTPS fallback bug fix
- RELEASE_NOTES: expand with EffectiveGitLabHosts design decisions,
env var merge semantics, and project-mode isolation
- Runbook: extend gitlab_hosts_config_runbook from 10 to 18 steps
covering env var merge, persistence safety, server parseOpts isolation,
and CLI-level env var smoke tests (18/18 PASS verified)
Self-managed GitLab instances on custom domains (e.g., git.company.com)
without "gitlab" in the hostname were parsed with 2-segment owner/repo
split instead of full-path repo. Users can now declare extra GitLab
hostnames in config so the parser treats them like gitlab.com (nested
subgroup support).
- Add `gitlab_hosts` field to both global and project Config structs
- Validate entries at load time (reject scheme, path, port, empty)
- Add `ParseOptions` / `ParseSourceWithOptions` to thread hosts into parser
- Add `isGitLabHost()` helper combining built-in + custom host detection
- Extend `InstallContext` interface with `GitLabHosts()` method
- Update all ParseSource call sites (install, search, update, server)
- Add JSON schemas for both config files
- Add unit tests for config validation and source parsing
- Add E2E runbook for devcontainer verification
- Document in configuration reference and URL formats appendix
The initial GitLab subgroup fix treated all multi-segment HTTPS paths as
full repo paths, regressing GHE and generic host subdir installs like
github.mycompany.com/org/repo/path/to/skill.
Add host-based heuristic: hosts containing "gitlab" use full-path-as-repo
behavior; all other hosts retain the original 2-segment owner/repo split
with remainder as subdir. This preserves backward compatibility for GHE,
Gitea, Gogs, and other non-GitLab platforms while correctly handling
GitLab nested subgroups.
Replace rigid 2-segment HTTPS regex with flexible path parsing that
treats the entire URL path as a repo when no explicit boundary exists.
Use .git suffix as the canonical repo/subdir delimiter, preserving
/-/ (GitLab) and /src/ (Bitbucket) web URL markers.
This fixes install failures for URLs like
onprem.gitlab.internal/org-group/sub1/sub2/project which previously
misinterpreted extra path segments as monorepo subdirectories.
The in-tree runbook runner (tools/runbook/, 25 Go files, ~4500 LOC)
has been extracted into the standalone mdproof project and is no
longer needed here.
Removed:
- tools/runbook/ — parser, executor, assertions, reporters, tests
- .devcontainer/bin/runbook — wrapper that built from tools/runbook
- ai_docs/runbook-guide.md — docs describing the old runner
- Makefile build-runbook target
The cli-e2e-test skill, runbook.json, and ensure-mdproof.sh already
reference the external mdproof binary for all E2E test execution.
Add a build hook that runs once before all runbooks, enabling
automatic binary rebuild. Lifecycle is now:
build (once) → [per runbook: setup → steps → teardown]
Build hook uses os/exec directly (not session executor), outputs
to stderr, and aborts immediately on failure.
Fix all 17 runbooks to pass in devcontainer:
- Adapt to session executor's set -eo pipefail (use cmd || EXIT=$?
pattern for expected non-zero exits)
- Replace embedded ```bash in heredocs with printf to avoid
confusing the runbook parser
- Fix negated assertion gotchas (Not FAIL, No such file)
- Remove fragile verify code blocks (multi-block merge issues)
- Fix step number collision (7b → 8, renumber subsequent)
- Use -s flag for batch installs instead of repeated --force
- Add (?m) flag for multi-line regex assertions
Add setup and teardown commands that run before/after each runbook.
Config file (runbook.json) provides persistent defaults; CLI flags
override. Setup env vars persist to all steps via session executor.
Setup failure skips all steps; teardown failure is informational only.
Removed Step 0 from atomgit runbook — init is now in runbook.json.
Auto-builds from source on each invocation, same pattern as the
ss/skillshare wrapper. Added to help screen and setup.sh quick
start message. Updated guide to document the shortcut.
Convert Expected blocks from natural language descriptions to
matchable assertions: exit_code: N, regex: <pattern>, jq: <expr>,
Not <text>, and plain substring matches. All runbooks now parse
correctly with the assertion engine.
SKILL.md now instructs AI to use bin/runbook --report json for
structured test execution. atomgit_install_runbook.md rewritten with
typed assertions (exit_code:, regex:, Not) as proof-of-concept.
JSON mode stdout was already pure JSON, but stderr still emitted
progress lines (e.g. "Discovering skills...", "Found N local skill(s)")
which broke `2>&1 | jq .` pipelines. Remove the else-branch stderr
writes so --json mode is fully silent except for stdout JSON output.
pterm's cursor package writes hide/show ANSI codes (\x1b[?25l/h)
directly to os.Stdout, bypassing WithWriter(). The previous fix
(SetProgressWriter to stderr) was insufficient because pterm still
created progress bar objects that triggered cursor.Hide()/Show().
Add SuppressProgressToStderr() which both redirects ProgressWriter
and forces isProgressTTY() to return false, preventing pterm
components from being instantiated at all. Extract assertNoANSI
test helper and cover all three structured formats (JSON, SARIF,
Markdown).
Also update pre-commit E2E runbook with steps 10-11 (framework
integration tests) and add pre-commit-hook recipe to sidebar.
- Add ai_docs/tests/precommit_audit_hook_runbook.md (9-step E2E test
for skillshare audit -p as pre-commit hook)
- Fix cli-e2e-test SKILL.md: clarify init -p has separate flag set from
global init; add audit rule ID vs pattern name guidance; add project
mode init example to container templates; add two new checklist items
- Update CHANGELOG.md with pre-commit hook, audit TUI filter, O(N)
cross-skill rewrite, and SARIF line-0 fix entries
filepath.Walk/WalkDir silently skip symlinked root directories because
os.Lstat sees the root as a non-directory symlink and never enters it.
This affected update, uninstall, reconcile, and server handlers when
the source directory was a symlink (common with dotfiles managers).
Extract ResolveSymlink() from sync.go into internal/utils/path.go and
apply it at every filepath.Walk/WalkDir callsite on source/target dirs.
Also add containment guards to resolveGroupSkills (uninstall) and
resolveGroupUpdatable (update) to reject group directories that are
symlinks pointing outside the source tree.
Covers 13 test scenarios for dotfiles manager symlink support:
symlinked source/target, chained symlinks, merge/copy modes,
status/diff/list/collect through symlinks, and idempotency.
UpdateGitIgnore/RemoveFromGitIgnore were called per-skill inside loops,
re-reading the entire .gitignore each time. With large files (100K+ lines
in project mode), this caused O(N×M) complexity and process hangs.
- Add UpdateGitIgnoreBatch and RemoveFromGitIgnoreBatch (single read/write)
- ReconcileProjectSkills: collect entries during walk, batch-update once
- uninstall (global + project): batch-remove after loop, not per-iteration
- Fix spinner WithRemoveWhenDone in StartSpinnerWithSteps
- Add E2E runbook for gitignore batch performance validation
- Change universal description from "(recommended)" to "(shared agent directory)"
in detect, checklist, and discover flows
- Add FAQ section on using universal alongside npx skills: merge vs copy mode
differences, name collision caveats, lock file visibility
- Add universal info admonition to supported-targets page
- Add E2E runbook for universal target path verification
- Update CHANGELOG and release notes for v0.16.4
Azure DevOps uses 4-segment paths (org/project/_git/repo) which the
existing 3-segment gitHTTPSPattern misparses. Add dedicated patterns
for modern HTTPS (dev.azure.com), legacy HTTPS (visualstudio.com),
and SSH v3 (ssh.dev.azure.com) formats. Also add ado: shorthand
(e.g. ado:org/project/repo) for quick installs.
Includes TrackName support (org-proj-repo format), 16 new test cases,
and url-formats.md documentation with platform support table update.
- 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
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
- 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)
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.
Add audit_output_antigravity_runbook.md covering install --all and
update --all audit output verification against sickn33/antigravity-
awesome-skills (7 steps, all passing).
Fix audit_output_online_test.go: batch install exits 0 when some
skills succeed (blocked count is a warning, not an error), so
AssertFailure should be AssertSuccess.
UpdateSkillsFromRepo was using filepath.Join(sourceDir, meta.Subdir) to
derive the local destination, but meta.Subdir is the repo-internal path
(e.g. "skills/foo") which may differ from the local install path ("foo/").
This created leaked copies at sourceDir/skills/foo/ alongside the correct
sourceDir/foo/, doubling the skill count on subsequent scans.
Change skillTargets from []string to map[string]string where key is the
repo subdir (for discovery lookup) and value is the local absolute path
(provided by the caller who already knows where the skill is installed).
file:// URLs with subdirectory paths (e.g., file:///repo/skills/alpha)
were passed whole to git clone, which fails because the full path isn't
a git repo. Add // separator support matching the existing SSH convention:
file:///path/to/repo//skills/alpha → CloneURL=file:///path/to/repo, Subdir=skills/alpha
Also updates E2E runbook Steps 10-11 to use the // syntax and improves
cli-e2e-test skill with cleanup prompt, cd /workspace note, and Go test template.
Implement remaining Issue 46 optimizations and developer experience fixes.
Install and update optimization:
- optimize install --track for subdirectory sources by attempting sparse checkout first while preserving git metadata
- fallback to existing tracked clone strategy on sparse capability or sparse path errors
- clean sparse created destination before fallback clone to avoid destination already exists failures
- improve tracked install spinner wording for subdir sources
- add tracked subdir unit tests for sparse success and sparse to full clone fallback
Progress user experience:
- normalize git progress lines by stripping volatile transfer rate suffix
- dedupe and throttle spinner and tree spinner update frequency to reduce tty flicker
- add unit tests for spinner progress normalization and throttling behavior
GitHub API messaging:
- improve rate limit guidance to suggest setting token env vars or using gh auth login
Devcontainer isolation:
- make ssenv enter command mode run from isolated environment cwd
- fix setup global init detection to use skillshare status global mode
- add auto mode hint in devcontainer wrapper when workspace project config takes precedence
E2E and tests:
- expand Issue 46 runbook with install, install track, update, no token fallback, sparse paths, and isolation checks
- update compact warning output test expectations to match current grouped warning rendering
Verification:
- go test internal install package
- go test internal install sparse and tracked subdir tests
- go test internal ui package
- go test cmd skillshare package
- go build skillshare binary and run integration TestInstall_Subdir
- run tracked subdir tests inside devcontainer
Move 5 skills from .claude/commands/ to .skillshare/skills/ so contributors
can share them via `skillshare sync -p`. Each skill has clear responsibility:
- cli-e2e-test: run isolated E2E tests in devcontainer
- codebase-audit: read-only cross-validation of code/docs/tests
- implement-feature: TDD workflow with E2E runbook generation
- update-docs: sync website docs with code changes
- changelog: generate CHANGELOG entries (RELEASE_NOTES maintainer-only)
Also:
- .gitignore: allow .skillshare/skills/, config.yaml, .gitignore to be tracked
- .skillignore: block .skillshare/skills/* from external install discovery
- ai_docs/tests/README.md: content merged into cli-e2e-test skill
Support bulk uninstall via 'skillshare uninstall --all' in both global
and project modes. Uses glob-based discovery to find all skills and
tracked repos, with --dry-run preview and --force to skip confirmation.
- Parse --all with mutual exclusion vs skill names and --group
- Global: discover skills + tracked repos from source directory
- Project: discover from .skillshare/skills/ directory
- Prune orphan symlinks from targets after removal
- Integration tests for --all, --all --dry-run, and conflict errors
- Update uninstall docs with --all examples
- Add AI runbook for uninstall --all glob testing
- Extend .skillshare-manifest.json tracking from copy mode to merge mode.
- Update PruneOrphanLinks to prioritize manifest records over naming patterns.
- Ensure SyncTargetMerge writes to manifest for all linked/updated skills.
- Add integration tests for uninstall + sync orphan directory cleanup.
- Maintain backward compatibility for targets without existing manifests.