[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79PP0Z042TZSZ71TMKXPRDE":3,"skill-same:01M79PP0Z042TZSZ71TMKXPRDE":57},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":13,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M79PP0Z042TZSZ71TMKXPRDE","starting-work",null,"Sync local main, then create a properly named branch (or worktree) to start new work. Use when starting new work, asked to \"start a new branch\", \"create a branch for X\", \"sync main and branch off\", or \"set up a worktree for this feature\" — before any commits happen. Validates the branch name against git-kit's [type]\u002F[description] convention and offers a worktree as an alternative to a plain branch checkout.","AndreHahm\u002Fandres-cc-marketplace","AndreHahm","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F80748019?v=4","01M79N2VZJT9MAJ7K695C9VXQ4","01M79N2W06PFX8BSCQ592M8K8K",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fstarting-work\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: starting-work\ndescription: >-\n  Sync local main, then create a properly named branch (or worktree) to start new work. Use when\n  starting new work, asked to \"start a new branch\", \"create a branch for X\", \"sync main and branch\n  off\", or \"set up a worktree for this feature\" — before any commits happen. Validates the branch name\n  against git-kit's [type]\u002F[description] convention and offers a worktree as an alternative to a plain\n  branch checkout.\nargument-hint: (optional) branch type and description, e.g. \"feature add-user-auth\"\nallowed-tools: Bash(git fetch:*), Bash(git checkout:*), Bash(git pull:*), Bash(git status:*), Bash(git branch --show-current:*), Bash(git symbolic-ref refs\u002Fremotes\u002Forigin\u002FHEAD:*), Bash(git worktree add:*), Bash(git worktree lock:*), Bash(bash ${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgit-write-marker.sh:*), Read, Glob, Write, AskUserQuestion\n---\n\n# Starting Work\n\nGet from \"I'm about to start something new\" to a properly named, up-to-date branch (or worktree) in one\ngated flow — instead of manually remembering to sync main, checking the naming convention, and deciding\nwhether a worktree makes sense, separately, every time.\n\n## When to Use\n\nBefore writing any code for a new piece of work: syncing local `main` and creating the branch (or\nworktree) it'll happen on. Triggers: \"start a new branch\", \"create a branch for X\", \"sync main and\nbranch off\", \"set up a worktree for this feature\".\n\n## When NOT to Use\n\n- **Rebasing an already-existing feature branch onto a newer main** — that's `\u002Fgit-sync-branch`, not this\n  skill. `starting-work` only runs before a branch exists.\n- **Renaming an existing branch based on what it now contains** — that's `\u002Fgit-update-branch-name`.\n- **Anything after the first commit** — once work has started, this skill has nothing left to do.\n\n## Instructions\n\n1. **Note the starting point and sync**: `git branch --show-current` (for the report). `git fetch origin --prune`.\n   Resolve the actual default branch — `git symbolic-ref refs\u002Fremotes\u002Forigin\u002FHEAD` (falling back to\n   `main` if that fails, e.g. no `origin` remote configured) — the same resolution `finishing-work` uses,\n   so the two halves of the branch lifecycle agree on what \"main\" means in a repo that defaults to\n   something else. Note: because the resolved branch name is dynamic, `allowed-tools` grants\n   `Bash(git checkout:*)` broadly rather than pinned to a literal branch name — this also covers step 4's\n   `git checkout -b`, so no separate grant is needed there. This skill never runs the file-restore form\n   (`git checkout -- \u003Cpath>`); only branch checkout and `-b` are ever invoked. If not already on the resolved default branch, ask via `AskUserQuestion` — \"Base the\n   new branch off `\u003Cresolved-branch>`, or off the current branch (`\u003Cname>`)?\" — default the resolved\n   branch. If basing off it: `git checkout \u003Cresolved-branch>`, then `git pull --ff-only`. If that fails\n   (diverged local branch, or it's already checked out in another worktree), tell the user exactly why\n   and stop rather than force anything. If basing off the current branch, skip the checkout\u002Fpull — see\n   `references\u002Fworktree-decision.md` for why this branch-off-a-branch case is legitimate (e.g. stacking\n   related work).\n   Verify `git status --porcelain` is clean before proceeding; if not, tell the user to commit or stash\n   first (point at the `commit` skill) and stop here.\n2. **Resolve the branch name**: use `$ARGUMENTS` if it supplies a type and description; otherwise ask\n   via `AskUserQuestion` for both, one at a time. Validate:\n   - Type: a single lowercase word, no slashes or spaces.\n   - Description: kebab-case — lowercase letters, digits, and hyphens only; no leading\u002Ftrailing\u002Fdouble\n     hyphen.\n   Don't hardcode or restate the accepted type list here — `commit`'s own \"Branch Naming Convention\"\n   section is the single source of truth for that. If the user is unsure what type to use, point them\n   there instead of guessing or duplicating the list. **If either value fails validation**, tell the\n   user why and ask again via `AskUserQuestion` — never pass an invalid type or description forward to\n   step 4, the same \"stop and explain\" discipline step 1 already applies to a diverged main.\n3. **Ask worktree or plain branch**: read `use_worktree` (default `true`) the same way `commit` reads its\n   own settings — `.claude\u002Fgit-kit.local.json` if it exists and sets the field, else the git-tracked\n   `${CLAUDE_PLUGIN_ROOT}\u002Fgit-kit.settings.json` default. This field doesn't need the trust-boundary check\n   `commit`'s `commit_confirm_before_commit`\u002F`commit_auto_stage` require — it never skips the question\n   below or triggers any automation on its own, it only changes which option is pre-highlighted, so honor\n   it from either file, tracked or not. Then always ask via `AskUserQuestion` — \"Create a plain branch, or\n   a separate worktree for this?\" — options \"Plain branch\" and \"Worktree\", with whichever `use_worktree`\n   favors listed first and suffixed `(Recommended)`: \"Worktree (Recommended)\" when `true` (the default,\n   chosen with multiple agents — e.g. Codex CLI alongside Claude Code — potentially working in this same\n   repo, where a separate working directory per piece of work avoids one agent's uncommitted state\n   colliding with another's), \"Plain branch (Recommended)\" when `false`. The setting never replaces this\n   ask — a human always makes the actual choice. See `references\u002Fworktree-decision.md` for the tradeoffs\n   to mention if the user wants guidance rather than a snap decision.\n4. **Create**: immediately before either branch-creating command below, run\n   `bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgit-write-marker.sh\" git-branch-create starting-work` — this writes\n   the marker git-kit's branch-creation guard requires; it must be written right before the command runs,\n   not earlier, since the hook only accepts a marker up to 60 seconds old.\n   - Plain branch: `git checkout -b \u003Ctype>\u002F\u003Cdescription>`.\n   - Worktree: compute a default path at `.claude\u002Fworktrees\u002F\u003Cdescription>` (this skill only ever runs\n     under Claude Code; the equivalent path for a Codex CLI session would be `.codex\u002Fworktrees\u002F\u003Cdescription>`\n     — both are gitignored). This is `starting-work`'s own default for a session-lifecycle-managed\n     worktree — distinct from `git-worktrees`' sibling-directory convention (`..\u002Fproject-feature`), which\n     stays the right default for manual, ad-hoc worktree management outside this automated flow; see that\n     skill's own note on the two use cases. Confirm the computed path with its own `AskUserQuestion` —\n     \"Create the worktree at `\u003Cpath>`?\" — options \"Use this path\" \u002F \"Enter a different path\". Don't rely\n     on step 3's worktree-vs-branch question to also cover the path (its own options only ever say\n     \"Plain branch\" \u002F \"Worktree\" — never fold the computed path into that option's description text as a\n     substitute for asking here). If the user supplies an override, constrain\n     it: letters, digits, hyphens, underscores, forward slashes, and periods only (no shell\n     metacharacters), it must not contain a `..` path segment, and it must resolve to a location either\n     (a) inside this repo under `.claude\u002Fworktrees\u002F` or `.codex\u002Fworktrees\u002F`, or (b) alongside\u002Fbelow the\n     repo's parent directory (the old sibling convention, still supported as an explicit override) — not\n     step 2's kebab-case rule, which is stricter than a filesystem path needs to be. If the override\n     doesn't meet this, tell the user why and ask again rather than passing it through. Then\n     `git worktree add -b \u003Ctype>\u002F\u003Cdescription> \u003Cpath>`, followed immediately by\n     `git worktree lock \u003Cpath> --reason \"claude session\"` so the worktree can't be removed by\n     `git worktree remove` (without `--force`) until it's explicitly unlocked — `git-cleanup` knows to\n     unlock a session-locked worktree before removing it once its branch is safe to delete.\n     Then copy over local, gitignored config `git worktree add` doesn't carry (it only shares tracked\n     history, not untracked files): `Glob` the main worktree root for `.claude\u002F*.local.json` (top-level\n     only, not recursive) and copy each match found into the same relative path in the new worktree,\n     only if the destination doesn't already exist there (never clobber). This covers\n     `.claude\u002Fsettings.local.json` plus every other per-plugin `.claude\u002F\u003Cname>.local.json` override this\n     repo's local-config precedent (`.claude\u002Frules\u002Fask-before-config-decisions.md`) produces — e.g.\n     `git-kit.local.json`, `analysis-kit.local.json`, `plugin-auditor.local.json`,\n     `codex-windows-guardrails.local.json` — generically, rather than a hardcoded list that needs a\n     follow-up fix every time a new plugin adopts the same precedent. Also `Glob` the main worktree for `**\u002FCLAUDE.local.md` and copy each\n     match found into the same relative path in the new worktree, same no-clobber rule — but first drop\n     any match under a vendored\u002Fthird-party tree (`node_modules\u002F`, `.venv\u002F`), this repo's own `.temp\u002F`\n     directory, or another worktree's own directory (`.claude\u002Fworktrees\u002F*\u002F`, `.codex\u002Fworktrees\u002F*\u002F`) —\n     this last exclusion matters because this same skill's own default worktree path lives under one of\n     those trees, so without it a `CLAUDE.local.md` belonging to a different, unrelated worktree could be\n     swept in and copied into the new one. Since a bare `**\u002FCLAUDE.local.md` pattern would otherwise\n     sweep all of these, and `Glob` itself has no exclude syntax, filter the match list before copying\n     rather than trying to encode the exclusion into the pattern.\n5. **Report**: the branch (or worktree path) just created, current location, and — for a worktree —\n   that `cd`-ing into the worktree path is needed before working there, and that it's now locked to this\n   session: when the work here is done, `finishing-work`\u002F`\u002Fgit-cleanup` handles unlocking and removal —\n   don't `git worktree remove --force` it directly, since that bypasses the lock as a safety signal that\n   this worktree is still in active use. Mention `git-worktrees` has further operations (compare, merge,\n   cleanup) if multiple worktrees end up in play.\n\n## Worktree Topic Scope\n\nA worktree this skill creates is scoped to one topic. If a session's already-bound worktree was created\nfor a different piece of work, don't start unrelated work in it — run this skill again for the new topic,\nwhich creates its own worktree. Symmetrically, don't keep working across worktrees for unrelated topics in\nthe same session: finish the current topic (`finishing-work` \u002F `\u002Fgit-cleanup`) before starting the next\none, rather than leaving multiple topics open at once.\n\nThis is a convention for how these worktrees get used, not a mechanical lock — nothing enforces \"topic\"\nthe way a file lock enforces exclusive access, since there's no way to check whether two pieces of work\nare actually the same topic from git state alone.\n\n**Why it matters:** a worktree that accumulates unrelated work becomes hard for `git-cleanup` to reason\nabout — it can't tell whether *all* the work in it is done, just the piece currently top of mind, so a\nmixed-topic worktree is more likely to get force-deleted along with unrelated in-progress changes by\nmistake. One worktree per topic keeps `git-cleanup`'s merged\u002Funmerged\u002Fdirty analysis meaningful: \"this\nworktree's branch merged\" reliably means \"this worktree is done.\"\n\nThe session-level half — don't leave one topic's worktree open while starting a second, unrelated one —\nmatters because the Stop exit-guard (`git-guard-dirty-worktree-exit.sh`) only checks the worktree the\nsession's current working directory is actually in when the turn ends, not every worktree the session has\never locked. Moving into a second worktree before finishing the first leaves the first one's dirty\u002Funmerged\nstate unchecked at Stop time, even though it's still locked and still at risk.\n\n## Testing & Validation\n\n**Verify this skill activates on:**\n- \"start a new branch for the auth refactor\"\n- \"create a branch for X\" \u002F \"sync main and branch off\"\n- \"set up a worktree for this feature\"\n\n**Verify it does NOT activate on:**\n- \"sync my current branch with main\" → `\u002Fgit-sync-branch`\n- \"rename this branch to match what it does now\" → `\u002Fgit-update-branch-name`\n- \"I just merged, clean this up\" → `finishing-work`\n\n**Verify `use_worktree` behavior:**\n- `use_worktree: true` (the default) — confirm step 3's `AskUserQuestion` still fires every time, with\n  \"Worktree (Recommended)\" listed first\n- `use_worktree: false` — confirm the same question fires, with \"Plain branch (Recommended)\" listed first\n  instead\n- Either value — confirm the setting never skips the question outright; the user's actual answer at step 3\n  always decides, regardless of which option was pre-highlighted\n\n**Verify the worktree-path confirmation:**\n- Worktree chosen at step 3, default path accepted at step 4's own `AskUserQuestion` — confirm\n  `git worktree add` runs with the computed default path\n- Worktree chosen at step 3, path overridden at step 4 — confirm the override goes through the existing\n  character-class\u002F`..`-segment\u002Flocation validation before being passed to `git worktree add`\n- Confirm the computed path is never mentioned inside step 3's own option description text as a substitute\n  for step 4's own question — step 3 asks *whether*, step 4 asks *where*\n\n**Verify the local-config copy behavior (issue #113):**\n- Main worktree root has `.claude\u002Fsettings.local.json` plus one or more other `.claude\u002F\u003Cname>.local.json`\n  files (e.g. `.claude\u002Fgit-kit.local.json`) — confirm every one of them is copied into the new worktree,\n  not just `settings.local.json`\n- A destination `.claude\u002F\u003Cname>.local.json` already exists in the new worktree — confirm it's never\n  clobbered\n- `.claude\u002F*.local.json` glob stays top-level only — confirm a `.local.json` file nested deeper than\n  `.claude\u002F` (not a realistic case today, but the pattern shouldn't accidentally reach one) is not swept\n\n**Verify the CLAUDE.local.md exclusion behavior:**\n- An unrelated, already-existing worktree (e.g. `.claude\u002Fworktrees\u002Fother-topic\u002F`) has its own\n  `CLAUDE.local.md` — confirm creating a new worktree never copies that file in; the sweep must drop\n  `.claude\u002Fworktrees\u002F*\u002F` and `.codex\u002Fworktrees\u002F*\u002F` matches the same way it already drops\n  `node_modules\u002F`\u002F`.venv\u002F`\u002F`.temp\u002F`\n- A `CLAUDE.local.md` genuinely at or below the main worktree root (not inside another worktree) is\n  still copied normally\n\n**Verified live, 2026-08-28:** ran `find .claude -maxdepth 1 -name \"*.local.json\"` against this repo's\nreal main checkout — matched exactly `.claude\u002Fsettings.local.json` and\n`.claude\u002Fcodex-windows-guardrails.local.json` (the same file issue #113's own reproduction steps cite),\nconfirming a `.claude\u002F*.local.json` glob reaches every real per-plugin local override this repo\ncurrently has, not just `settings.local.json`. No fresh `skill-tester` eval re-run for this edit — the\nfix is a mechanical glob-pattern generalization of already-eval-covered copy logic\n(`evals\u002Fstarting-work\u002Fevals.json`), verified directly against real repo files rather than re-run\nend-to-end.\n\n**Quality gates:**\n- [ ] Step 1 never fast-forwards a diverged local `main` silently — always stops and tells the user\n- [ ] Step 2 never hardcodes the branch-type list — always points at `commit`'s own convention section\n- [ ] Step 2 never passes an invalid type\u002Fdescription forward — always re-asks on validation failure\n- [ ] Step 3's worktree-vs-branch question always uses `AskUserQuestion`, never assumed — `use_worktree`\n      only changes which option is recommended, it never skips the question\n- [ ] Step 4 never passes an unconstrained worktree-path override to `git worktree add` — always\n      validates the character class and rejects any `..` path segment\n- [ ] Step 4 always fires its own `AskUserQuestion` for the worktree path before `git worktree add` runs —\n      a path mentioned only inside step 3's option description never satisfies this\n- [ ] Step 4's default worktree path is always `.claude\u002Fworktrees\u002F\u003Cdescription>` (or `.codex\u002Fworktrees\u002F`\n      for a Codex CLI session) — never the old sibling-directory convention, unless the user explicitly\n      overrides to one\n- [ ] Every worktree Step 4 creates is locked (`git worktree lock`) immediately after `git worktree add`\n      — never left unlocked\n- [ ] Step 4's local-config copy always globs `.claude\u002F*.local.json` (every per-plugin override), never\n      just `.claude\u002Fsettings.local.json` alone\n- [ ] Step 4's `CLAUDE.local.md` sweep always drops matches under another worktree's own directory\n      (`.claude\u002Fworktrees\u002F*\u002F`, `.codex\u002Fworktrees\u002F*\u002F`), never just `node_modules\u002F`\u002F`.venv\u002F`\u002F`.temp\u002F`\n- [ ] Step 5's report always mentions the worktree is session-locked and points at\n      `finishing-work`\u002F`\u002Fgit-cleanup` for removal — never suggests `git worktree remove --force` directly\n- [ ] Step 4 always writes the `git-branch-create` marker immediately before `git checkout -b` \u002F\n      `git worktree add -b`, never earlier in the run\n- [ ] A dirty working tree at step 1 always stops the flow with a pointer to `commit`, never proceeds\n\n## Reference Guide\n\n| Resource | Purpose |\n|---|---|\n| `references\u002Fworktree-decision.md` | Worktree-vs-plain-branch tradeoffs, and the branch-off-a-branch escape hatch for stacking work on top of an open PR |\n","# Starting Work\n\nGet from \"I'm about to start something new\" to a properly named, up-to-date branch (or worktree) in one\ngated flow — instead of manually remembering to sync main, checking the naming convention, and deciding\nwhether a worktree makes sense, separately, every time.\n\n## When to Use\n\nBefore writing any code for a new piece of work: syncing local `main` and creating the branch (or\nworktree) it'll happen on. Triggers: \"start a new branch\", \"create a branch for X\", \"sync main and\nbranch off\", \"set up a worktree for this feature\".\n\n## When NOT to Use\n\n- **Rebasing an already-existing feature branch onto a newer main** — that's `\u002Fgit-sync-branch`, not this\n  skill. `starting-work` only runs before a branch exists.\n- **Renaming an existing branch based on what it now contains** — that's `\u002Fgit-update-branch-name`.\n- **Anything after the first commit** — once work has started, this skill has nothing left to do.\n\n## Instructions\n\n1. **Note the starting point and sync**: `git branch --show-current` (for the report). `git fetch origin --prune`.\n   Resolve the actual default branch — `git symbolic-ref refs\u002Fremotes\u002Forigin\u002FHEAD` (falling back to\n   `main` if that fails, e.g. no `origin` remote configured) — the same resolution `finishing-work` uses,\n   so the two halves of the branch lifecycle agree on what \"main\" means in a repo that defaults to\n   something else. Note: because the resolved branch name is dynamic, `allowed-tools` grants\n   `Bash(git checkout:*)` broadly rather than pinned to a literal branch name — this also covers step 4's\n   `git checkout -b`, so no separate grant is needed there. This skill never runs the file-restore form\n   (`git checkout -- \u003Cpath>`); only branch checkout and `-b` are ever invoked. If not already on the resolved default branch, ask via `AskUserQuestion` — \"Base the\n   new branch off `\u003Cresolved-branch>`, or off the current branch (`\u003Cname>`)?\" — default the resolved\n   branch. If basing off it: `git checkout \u003Cresolved-branch>`, then `git pull --ff-only`. If that fails\n   (diverged local branch, or it's already checked out in another worktree), tell the user exactly why\n   and stop rather than force anything. If basing off the current branch, skip the checkout\u002Fpull — see\n   `references\u002Fworktree-decision.md` for why this branch-off-a-branch case is legitimate (e.g. stacking\n   related work).\n   Verify `git status --porcelain` is clean before proceeding; if not, tell the user to commit or stash\n   first (point at the `commit` skill) and stop here.\n2. **Resolve the branch name**: use `$ARGUMENTS` if it supplies a type and description; otherwise ask\n   via `AskUserQuestion` for both, one at a time. Validate:\n   - Type: a single lowercase word, no slashes or spaces.\n   - Description: kebab-case — lowercase letters, digits, and hyphens only; no leading\u002Ftrailing\u002Fdouble\n     hyphen.\n   Don't hardcode or restate the accepted type list here — `commit`'s own \"Branch Naming Convention\"\n   section is the single source of truth for that. If the user is unsure what type to use, point them\n   there instead of guessing or duplicating the list. **If either value fails validation**, tell the\n   user why and ask again via `AskUserQuestion` — never pass an invalid type or description forward to\n   step 4, the same \"stop and explain\" discipline step 1 already applies to a diverged main.\n3. **Ask worktree or plain branch**: read `use_worktree` (default `true`) the same way `commit` reads its\n   own settings — `.claude\u002Fgit-kit.local.json` if it exists and sets the field, else the git-tracked\n   `${CLAUDE_PLUGIN_ROOT}\u002Fgit-kit.settings.json` default. This field doesn't need the trust-boundary check\n   `commit`'s `commit_confirm_before_commit`\u002F`commit_auto_stage` require — it never skips the question\n   below or triggers any automation on its own, it only changes which option is pre-highlighted, so honor\n   it from either file, tracked or not. Then always ask via `AskUserQuestion` — \"Create a plain branch, or\n   a separate worktree for this?\" — options \"Plain branch\" and \"Worktree\", with whichever `use_worktree`\n   favors listed first and suffixed `(Recommended)`: \"Worktree (Recommended)\" when `true` (the default,\n   chosen with multiple agents — e.g. Codex CLI alongside Claude Code — potentially working in this same\n   repo, where a separate working directory per piece of work avoids one agent's uncommitted state\n   colliding with another's), \"Plain branch (Recommended)\" when `false`. The setting never replaces this\n   ask — a human always makes the actual choice. See `references\u002Fworktree-decision.md` for the tradeoffs\n   to mention if the user wants guidance rather than a snap decision.\n4. **Create**: immediately before either branch-creating command below, run\n   `bash \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgit-write-marker.sh\" git-branch-create starting-work` — this writes\n   the marker git-kit's branch-creation guard requires; it must be written right before the command runs,\n   not earlier, since the hook only accepts a marker up to 60 seconds old.\n   - Plain branch: `git checkout -b \u003Ctype>\u002F\u003Cdescription>`.\n   - Worktree: compute a default path at `.claude\u002Fworktrees\u002F\u003Cdescription>` (this skill only ever runs\n     under Claude Code; the equivalent path for a Codex CLI session would be `.codex\u002Fworktrees\u002F\u003Cdescription>`\n     — both are gitignored). This is `starting-work`'s own default for a session-lifecycle-managed\n     worktree — distinct from `git-worktrees`' sibling-directory convention (`..\u002Fproject-feature`), which\n     stays the right default for manual, ad-hoc worktree management outside this automated flow; see that\n     skill's own note on the two use cases. Confirm the computed path with its own `AskUserQuestion` —\n     \"Create the worktree at `\u003Cpath>`?\" — options \"Use this path\" \u002F \"Enter a different path\". Don't rely\n     on step 3's worktree-vs-branch question to also cover the path (its own options only ever say\n     \"Plain branch\" \u002F \"Worktree\" — never fold the computed path into that option's description text as a\n     substitute for asking here). If the user supplies an override, constrain\n     it: letters, digits, hyphens, underscores, forward slashes, and periods only (no shell\n     metacharacters), it must not contain a `..` path segment, and it must resolve to a location either\n     (a) inside this repo under `.claude\u002Fworktrees\u002F` or `.codex\u002Fworktrees\u002F`, or (b) alongside\u002Fbelow the\n     repo's parent directory (the old sibling convention, still supported as an explicit override) — not\n     step 2's kebab-case rule, which is stricter than a filesystem path needs to be. If the override\n     doesn't meet this, tell the user why and ask again rather than passing it through. Then\n     `git worktree add -b \u003Ctype>\u002F\u003Cdescription> \u003Cpath>`, followed immediately by\n     `git worktree lock \u003Cpath> --reason \"claude session\"` so the worktree can't be removed by\n     `git worktree remove` (without `--force`) until it's explicitly unlocked — `git-cleanup` knows to\n     unlock a session-locked worktree before removing it once its branch is safe to delete.\n     Then copy over local, gitignored config `git worktree add` doesn't carry (it only shares tracked\n     history, not untracked files): `Glob` the main worktree root for `.claude\u002F*.local.json` (top-level\n     only, not recursive) and copy each match found into the same relative path in the new worktree,\n     only if the destination doesn't already exist there (never clobber). This covers\n     `.claude\u002Fsettings.local.json` plus every other per-plugin `.claude\u002F\u003Cname>.local.json` override this\n     repo's local-config precedent (`.claude\u002Frules\u002Fask-before-config-decisions.md`) produces — e.g.\n     `git-kit.local.json`, `analysis-kit.local.json`, `plugin-auditor.local.json`,\n     `codex-windows-guardrails.local.json` — generically, rather than a hardcoded list that needs a\n     follow-up fix every time a new plugin adopts the same precedent. Also `Glob` the main worktree for `**\u002FCLAUDE.local.md` and copy each\n     match found into the same relative path in the new worktree, same no-clobber rule — but first drop\n     any match under a vendored\u002Fthird-party tree (`node_modules\u002F`, `.venv\u002F`), this repo's own `.temp\u002F`\n     directory, or another worktree's own directory (`.claude\u002Fworktrees\u002F*\u002F`, `.codex\u002Fworktrees\u002F*\u002F`) —\n     this last exclusion matters because this same skill's own default worktree path lives under one of\n     those trees, so without it a `CLAUDE.local.md` belonging to a different, unrelated worktree could be\n     swept in and copied into the new one. Since a bare `**\u002FCLAUDE.local.md` pattern would otherwise\n     sweep all of these, and `Glob` itself has no exclude syntax, filter the match list before copying\n     rather than trying to encode the exclusion into the pattern.\n5. **Report**: the branch (or worktree path) just created, current location, and — for a worktree —\n   that `cd`-ing into the worktree path is needed before working there, and that it's now locked to this\n   session: when the work here is done, `finishing-work`\u002F`\u002Fgit-cleanup` handles unlocking and removal —\n   don't `git worktree remove --force` it directly, since that bypasses the lock as a safety signal that\n   this worktree is still in active use. Mention `git-worktrees` has further operations (compare, merge,\n   cleanup) if multiple worktrees end up in play.\n\n## Worktree Topic Scope\n\nA worktree this skill creates is scoped to one topic. If a session's already-bound worktree was created\nfor a different piece of work, don't start unrelated work in it — run this skill again for the new topic,\nwhich creates its own worktree. Symmetrically, don't keep working across worktrees for unrelated topics in\nthe same session: finish the current topic (`finishing-work` \u002F `\u002Fgit-cleanup`) before starting the next\none, rather than leaving multiple topics open at once.\n\nThis is a convention for how these worktrees get used, not a mechanical lock — nothing enforces \"topic\"\nthe way a file lock enforces exclusive access, since there's no way to check whether two pieces of work\nare actually the same topic from git state alone.\n\n**Why it matters:** a worktree that accumulates unrelated work becomes hard for `git-cleanup` to reason\nabout — it can't tell whether *all* the work in it is done, just the piece currently top of mind, so a\nmixed-topic worktree is more likely to get force-deleted along with unrelated in-progress changes by\nmistake. One worktree per topic keeps `git-cleanup`'s merged\u002Funmerged\u002Fdirty analysis meaningful: \"this\nworktree's branch merged\" reliably means \"this worktree is done.\"\n\nThe session-level half — don't leave one topic's worktree open while starting a second, unrelated one —\nmatters because the Stop exit-guard (`git-guard-dirty-worktree-exit.sh`) only checks the worktree the\nsession's current working directory is actually in when the turn ends, not every worktree the session has\never locked. Moving into a second worktree before finishing the first leaves the first one's dirty\u002Funmerged\nstate unchecked at Stop time, even though it's still locked and still at risk.\n\n## Testing & Validation\n\n**Verify this skill activates on:**\n- \"start a new branch for the auth refactor\"\n- \"create a branch for X\" \u002F \"sync main and branch off\"\n- \"set up a worktree for this feature\"\n\n**Verify it does NOT activate on:**\n- \"sync my current branch with main\" → `\u002Fgit-sync-branch`\n- \"rename this branch to match what it does now\" → `\u002Fgit-update-branch-name`\n- \"I just merged, clean this up\" → `finishing-work`\n\n**Verify `use_worktree` behavior:**\n- `use_worktree: true` (the default) — confirm step 3's `AskUserQuestion` still fires every time, with\n  \"Worktree (Recommended)\" listed first\n- `use_worktree: false` — confirm the same question fires, with \"Plain branch (Recommended)\" listed first\n  instead\n- Either value — confirm the setting never skips the question outright; the user's actual answer at step 3\n  always decides, regardless of which option was pre-highlighted\n\n**Verify the worktree-path confirmation:**\n- Worktree chosen at step 3, default path accepted at step 4's own `AskUserQuestion` — confirm\n  `git worktree add` runs with the computed default path\n- Worktree chosen at step 3, path overridden at step 4 — confirm the override goes through the existing\n  character-class\u002F`..`-segment\u002Flocation validation before being passed to `git worktree add`\n- Confirm the computed path is never mentioned inside step 3's own option description text as a substitute\n  for step 4's own question — step 3 asks *whether*, step 4 asks *where*\n\n**Verify the local-config copy behavior (issue #113):**\n- Main worktree root has `.claude\u002Fsettings.local.json` plus one or more other `.claude\u002F\u003Cname>.local.json`\n  files (e.g. `.claude\u002Fgit-kit.local.json`) — confirm every one of them is copied into the new worktree,\n  not just `settings.local.json`\n- A destination `.claude\u002F\u003Cname>.local.json` already exists in the new worktree — confirm it's never\n  clobbered\n- `.claude\u002F*.local.json` glob stays top-level only — confirm a `.local.json` file nested deeper than\n  `.claude\u002F` (not a realistic case today, but the pattern shouldn't accidentally reach one) is not swept\n\n**Verify the CLAUDE.local.md exclusion behavior:**\n- An unrelated, already-existing worktree (e.g. `.claude\u002Fworktrees\u002Fother-topic\u002F`) has its own\n  `CLAUDE.local.md` — confirm creating a new worktree never copies that file in; the sweep must drop\n  `.claude\u002Fworktrees\u002F*\u002F` and `.codex\u002Fworktrees\u002F*\u002F` matches the same way it already drops\n  `node_modules\u002F`\u002F`.venv\u002F`\u002F`.temp\u002F`\n- A `CLAUDE.local.md` genuinely at or below the main worktree root (not inside another worktree) is\n  still copied normally\n\n**Verified live, 2026-08-28:** ran `find .claude -maxdepth 1 -name \"*.local.json\"` against this repo's\nreal main checkout — matched exactly `.claude\u002Fsettings.local.json` and\n`.claude\u002Fcodex-windows-guardrails.local.json` (the same file issue #113's own reproduction steps cite),\nconfirming a `.claude\u002F*.local.json` glob reaches every real per-plugin local override this repo\ncurrently has, not just `settings.local.json`. No fresh `skill-tester` eval re-run for this edit — the\nfix is a mechanical glob-pattern generalization of already-eval-covered copy logic\n(`evals\u002Fstarting-work\u002Fevals.json`), verified directly against real repo files rather than re-run\nend-to-end.\n\n**Quality gates:**\n- [ ] Step 1 never fast-forwards a diverged local `main` silently — always stops and tells the user\n- [ ] Step 2 never hardcodes the branch-type list — always points at `commit`'s own convention section\n- [ ] Step 2 never passes an invalid type\u002Fdescription forward — always re-asks on validation failure\n- [ ] Step 3's worktree-vs-branch question always uses `AskUserQuestion`, never assumed — `use_worktree`\n      only changes which option is recommended, it never skips the question\n- [ ] Step 4 never passes an unconstrained worktree-path override to `git worktree add` — always\n      validates the character class and rejects any `..` path segment\n- [ ] Step 4 always fires its own `AskUserQuestion` for the worktree path before `git worktree add` runs —\n      a path mentioned only inside step 3's option description never satisfies this\n- [ ] Step 4's default worktree path is always `.claude\u002Fworktrees\u002F\u003Cdescription>` (or `.codex\u002Fworktrees\u002F`\n      for a Codex CLI session) — never the old sibling-directory convention, unless the user explicitly\n      overrides to one\n- [ ] Every worktree Step 4 creates is locked (`git worktree lock`) immediately after `git worktree add`\n      — never left unlocked\n- [ ] Step 4's local-config copy always globs `.claude\u002F*.local.json` (every per-plugin override), never\n      just `.claude\u002Fsettings.local.json` alone\n- [ ] Step 4's `CLAUDE.local.md` sweep always drops matches under another worktree's own directory\n      (`.claude\u002Fworktrees\u002F*\u002F`, `.codex\u002Fworktrees\u002F*\u002F`), never just `node_modules\u002F`\u002F`.venv\u002F`\u002F`.temp\u002F`\n- [ ] Step 5's report always mentions the worktree is session-locked and points at\n      `finishing-work`\u002F`\u002Fgit-cleanup` for removal — never suggests `git worktree remove --force` directly\n- [ ] Step 4 always writes the `git-branch-create` marker immediately before `git checkout -b` \u002F\n      `git worktree add -b`, never earlier in the run\n- [ ] A dirty working tree at step 1 always stops the flow with a pointer to `commit`, never proceeds\n\n## Reference Guide\n\n| Resource | Purpose |\n|---|---|\n| `references\u002Fworktree-decision.md` | Worktree-vs-plain-branch tradeoffs, and the branch-off-a-branch escape hatch for stacking work on top of an open PR |\n",{"name":5,"description":7,"allowed-tools":23,"argument-hint":24},"Bash(git fetch:*), Bash(git checkout:*), Bash(git pull:*), Bash(git status:*), Bash(git branch --show-current:*), Bash(git symbolic-ref refs\u002Fremotes\u002Forigin\u002FHEAD:*), Bash(git worktree add:*), Bash(git worktree lock:*), Bash(bash ${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fgit-write-marker.sh:*), Read, Glob, Write, AskUserQuestion","(optional) branch type and description, e.g. \"feature add-user-auth\"","main","1bf12900cac6f28e346abd2912bdb78d306de723",".claude\u002Fskills\u002Fstarting-work","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace","https:\u002F\u002Fgithub.com\u002FAndreHahm","2026-11-15 04:26:16",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add AndreHahm\u002Fandres-cc-marketplace --skill starting-work",{"tool":37,"command":38},"bunx","bunx skills add AndreHahm\u002Fandres-cc-marketplace --skill starting-work",{"tool":40,"command":41},"pnpm","pnpm dlx skills add AndreHahm\u002Fandres-cc-marketplace --skill starting-work","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fstarting-work\u002FSKILL.md",true,[45,51],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"references\u002Fworktree-decision.md","REFERENCE","参考文档",1442,"https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fstarting-work\u002Freferences\u002Fworktree-decision.md",{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"scripts\u002Fsmoke_test.py","SCRIPT","脚本",7966,"https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fstarting-work\u002Fscripts\u002Fsmoke_test.py",[58,65,72,79,86,93],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":18},"01M79R0K918Z0XDM74R0EY5H3P","demo","Fixture skill used by scripts.marketplace_ci sync tests.","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Ftests\u002Fmarketplace_ci\u002Ffixtures\u002Fplugins\u002Fsample-kit\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M79R08VMDJA2FPJKH3ZM309K","export-demo","Fixture skill used by scripts.marketplace_ci export tests.","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Ftests\u002Fmarketplace_ci\u002Ffixtures\u002Fclaude\u002Fskills\u002Fexport-demo\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M79R042FS581DS2JNW5AZ1MK","work-to-development","Validate an accepted Linear Issue's readiness for implementation (criteria, dependencies, duplicate\u002Fexisting artifacts) and request governed branch\u002Fworktree creation through git-kit:starting-work — recording work-started Git\u002FGitHub evidence and a deliberate Linear Started transition only after Git identity is read back and verified. Use when asked to start work on a Linear issue, begin development for an accepted issue, or prepare an accepted issue for implementation. Never creates a branch directly — always delegates to git-kit.","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fworkmanagement-kit\u002Fskills\u002Fwork-to-development\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M79QZZD4EHB82P72P4N92BKT","work-linking","Create and maintain stable cross-system links and authority labels between Notion and Linear records, and detect\u002Frepair drift between them. Use when asked to link a Notion record to a Linear record, check for drift between Notion and Linear, or repair a broken\u002Fstale link. Repairs only the bounded non-authoritative copy or link — never chooses by newest timestamp, and never creates a reverse-write loop against either system's own authority.","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fworkmanagement-kit\u002Fskills\u002Fwork-linking\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M79QZVFA3GTVNZMJ51NESVSP","status-and-learning","Produce a dated, explicitly non-live Linear-to-Notion progress summary, or capture an outcome\u002Flearning after work completes. Use when asked to summarize progress to Notion, report portfolio status, or capture what was learned from a completed piece of work — not to disposition open follow-up items (see open-item-management for that). Never runs automatically or continuously — every summary is a deliberate, approved snapshot, not a live sync.","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fworkmanagement-kit\u002Fskills\u002Fstatus-and-learning\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":18},"01M79QZNSKSABK7X2PT93X8QT5","repository-gates","Discover, delegate, record, and invalidate this repository's actual Git\u002FGitHub lifecycle gates (pre-commit, pre-push, PR required checks, Codex delta review, review requirements, merge rights) and the repository-policy provider profile that maps governed operations to git-kit. Use when asked what gates apply to this repository, to discover repository policy before starting Wave 2 work, or to check whether prior gate evidence is still valid after new commits. Discovery and delegation only — never executes a gate's own internal logic, and never invents a gate the repository doesn't actually define (in particular, never assumes a \"Review Changes\" gate exists).","https:\u002F\u002Fgithub.com\u002FAndreHahm\u002Fandres-cc-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fworkmanagement-kit\u002Fskills\u002Frepository-gates\u002FSKILL.md",[],[]]