[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79ZTSG7RHTA5BZPDSW8VWEN":3,"skill-same:01M79ZTSG7RHTA5BZPDSW8VWEN":61},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":34,"ownerDescription":35,"updatedAt":36,"install":37,"resources":50},"01M79ZTSG7RHTA5BZPDSW8VWEN","dispatch-resolve-wave",null,"Dispatch file-disjoint issue_resolver worktrees for the next resolve wave. Use when review_orchestrator is in the Review resolve loop, or when the user says \u002Fdispatch_resolve_wave. Do not implement fixes.","sazlin\u002Floadout","sazlin","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F7120031?v=4","01M79ZSNX5RMXATQTSH5NGHN51","Sean Azlin","01M79ZSNXQ5HCTYC1RE7J5CAPF",0,"MIT","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdispatch-resolve-wave\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: dispatch-resolve-wave\ndescription: Dispatch file-disjoint issue_resolver worktrees for the next resolve\n  wave. Use when review_orchestrator is in the Review resolve loop, or when the user\n  says \u002Fdispatch_resolve_wave. Do not implement fixes.\nmetadata:\n  loadout.managed: 'true'\n  loadout.source: skills\u002Fdispatch-resolve-wave\u002FSKILL.md\n  loadout.sha: 49eb9bd\n---\n\n# Dispatch resolve wave\n\nLaunch one `issue_resolver` per file-disjoint open task in the next wave,\n**in one turn, in parallel**. Do not implement fixes. Worktree git (add,\ncherry-pick, prune) is owned by `dispatch-resolve-wave` via\n`prepare_wave_worktrees.py`. The orchestrator keeps mark-done,\nlog-progress, and is the only `git push` of `origin\u002F\u003Cpr-head>`.\n\n## When to use\n\n- `review_orchestrator` is in the Review resolve loop\n- The user asks for `\u002Fdispatch_resolve_wave`\n\n**Do not use** to implement fixes, mark tasks done, or merge.\n\n## Steps\n\n1. Run `python3 .claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fpartition_waves.py \u003Ctasks_path>`.\n   If `.claude\u002F` is missing, use\n   `skills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fpartition_waves.py`. The CLI prints\n   `{\"wave\": [\"TASK-001\", ...]}`. Wave cap 4 comes from the script. Never\n   exceed the script's wave.\n2. Create one worktree and task branch per wave task from current PR HEAD.\n   Do not `git checkout` the task branch in the PR worktree.\n   Run `python3 .claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fprepare_wave_worktrees.py add --pr-head \u003Cref> \u003Cwave-json>`\n   (or `skills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fprepare_wave_worktrees.py` if\n   `.claude\u002F` is missing). Pass the PR head and wave JSON as argv to that\n   helper only — never interpolate PR head, task ids, or SHAs into a shell\n   git command. The helper validates the ref with `git check-ref-format\n   --normalize`, validates each id with `^TASK-[0-9]+$`, maps `\u002F` in the\n   branch name to a single `.worktrees\u002F` path segment, and calls\n   `git worktree add` \u002F `git branch` via `subprocess.run` list argv (no\n   `shell=True`). It reuses leftover dirs (`git reset --hard` to PR HEAD),\n   falls back to `\u002Ftmp\u002Fpr-resolve-\u003CTASK-ID>`, and creates the task branch\n   without moving PR HEAD when both add sites fail. Use the JSON\n   `worktree` \u002F `branch` fields; do not assemble those paths yourself.\n   Helper `add` JSON `isolated: true` means the checkout exists; it does not copy\n   `tasks_path` — the caller must copy it. After each successful\n   `git worktree add` (including the `\u002Ftmp` fallback) and after\n   reuse\u002Freset, copy the frozen gitignored\n   `tasks_path` from the PR worktree into that worktree (same filename)\n   so `test -f \u003Cworktree>\u002F\u003Ctasks_path>` is true and the bytes match the\n   PR-worktree manifest. A resolver restricted to the worktree must read\n   the assigned task without consulting the PR worktree. If copy is\n   skipped, pass an absolute PR-worktree `tasks_path` and say so in the\n   brief.\n3. If any wave task still has no isolated dir after reuse\u002Freset and both\n   add sites, do not stop the whole wave. Dispatch `issue_resolver` with\n   `resolve-next-task` for every task whose worktree succeeded. Leave ids\n   with no isolated dir `[open]` (undispatched). Sequential one-task mode\n   is not used solely because of leftovers.\n4. Otherwise — only when every wave task has an isolated checkout —\n   dispatch **all wave resolvers in one turn**, in parallel (same protocol\n   as panel). Issue the calls in a **single** response. One call per\n   response is a protocol failure.\n5. Each brief: `issue_resolver`, `resolve-next-task`, `tasks_path`,\n   `task_id`, worktree path, task branch, PR head branch name. Git cwd is\n   the worktree; `tasks_path` is the copied file in that worktree or the\n   absolute PR-worktree manifest path named in this brief. \"Do not push PR\n   head. Do not edit the tasks file or REVIEW_HISTORY.md.\"\n6. After all return: in TASK id order, run\n   `python3 ...\u002Fprepare_wave_worktrees.py cherry-pick --task-branch \u003Cbranch> [--] \u003Csha>`\n   for each resolver (the helper `git fetch`es the task branch, then\n   `git cherry-pick` with the SHA after `--`). Success per task id is\n   resolver `status=ok` plus a commit SHA that applies cleanly.\n   Missing SHA, blocked resolver, or status other than `ok`: skip\n   cherry-pick (no placeholder SHA); leave that id `[open]`. On any\n   non-zero cherry-pick (conflict, bad or missing object) the helper runs\n   `git cherry-pick --abort`; leave that id `[open]`, continue with the\n   next wave task. HEAD must not remain in a cherry-pick or merge state.\n   Do not push PR head. Undispatched fallback tasks stay `[open]`. Never\n   hand-build `git cherry-pick` \u002F `git fetch` lines from PR head or SHA\n   strings.\n7. After cherry-picks (success, conflict abort, or missing sha) and after\n   every wave (success, conflict, or dispatch failure), always run\n   `python3 ...\u002Fprepare_wave_worktrees.py prune --pr-head \u003Cref> \u003Cwave-json>`.\n   The helper runs `git worktree remove --force` on the wave paths,\n   `git worktree prune`, deletes local `\u003Cpr-head>-\u003CTASK-ID>` branches, and\n   removes leftover `\u002Ftmp\u002Fpr-resolve-\u003CTASK-ID>` dirs that are not\n   registered worktrees. Recreate them on the next attempt. On\n   abort-if-merged and on resolve-loop exit, prune any remaining\n   `.worktrees\u002F\u003Cpr-head>-*` paths and matching `\u002Ftmp` fallbacks\n   (worktree\u002Fbranch\u002Ftmp cleanup, not only child cancel\u002Farchive) before\n   emitting JSON. Never hand-build those git lines from PR head strings.\n8. After cherry-picks, stop. Never push `origin\u002F\u003Cpr-head>` in this skill.\n   Worktree git (add, cherry-pick, prune) is owned by `dispatch-resolve-wave`\n   via `prepare_wave_worktrees.py`. The orchestrator keeps mark-done,\n   log-progress, and is the only `git push` of `origin\u002F\u003Cpr-head>` — marks\n   `[done]` only the ids whose cherry-pick landed cleanly (not the whole wave),\n   follows `log-progress`, and is the only `git push` of `origin\u002F\u003Cpr-head>` —\n   and only after at least one clean pick and only if HEAD is not in a\n   cherry-pick or merge state.\n\n## Resolve-wave retry cap\n\nA wave with no successful cherry-pick, or a task that conflicted \u002F returned\n`blocked` \u002F produced no sha, is one failed attempt for those task ids.\nCap retries at **3** (short backoff), same as other failure classes.\nConflicted or blocked ids are not immediately eligible for another\nidentical parallel wave in the same run beyond this cap. After the cap,\nleave those tasks `[open]` and stop this skill; the orchestrator exits the\nresolve loop with `open_task_ids` and continues panel\u002Fverify\u002Frisk. Do not\ndispatch a fourth identical parallel wave for the same ids in this run.\n\n## Harness\n\n- Cursor: one `Task` call per wave task in a single message (`issue_resolver`).\n- Claude Code: one Agent call per wave task using `issue_resolver`.\n\n## Guardrails\n\n- Never push `origin\u002F\u003Cpr-head>` (orchestrator pushes after successful picks)\n- Never rebase\n- Never force-push\n- Never open a second PR\n- Never exceed the script's wave\n- Never implement fixes\n- Never interpolate PR head refs or commit SHAs into a shell git command;\n  run `prepare_wave_worktrees.py` instead\n","# Dispatch resolve wave\n\nLaunch one `issue_resolver` per file-disjoint open task in the next wave,\n**in one turn, in parallel**. Do not implement fixes. Worktree git (add,\ncherry-pick, prune) is owned by `dispatch-resolve-wave` via\n`prepare_wave_worktrees.py`. The orchestrator keeps mark-done,\nlog-progress, and is the only `git push` of `origin\u002F\u003Cpr-head>`.\n\n## When to use\n\n- `review_orchestrator` is in the Review resolve loop\n- The user asks for `\u002Fdispatch_resolve_wave`\n\n**Do not use** to implement fixes, mark tasks done, or merge.\n\n## Steps\n\n1. Run `python3 .claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fpartition_waves.py \u003Ctasks_path>`.\n   If `.claude\u002F` is missing, use\n   `skills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fpartition_waves.py`. The CLI prints\n   `{\"wave\": [\"TASK-001\", ...]}`. Wave cap 4 comes from the script. Never\n   exceed the script's wave.\n2. Create one worktree and task branch per wave task from current PR HEAD.\n   Do not `git checkout` the task branch in the PR worktree.\n   Run `python3 .claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fprepare_wave_worktrees.py add --pr-head \u003Cref> \u003Cwave-json>`\n   (or `skills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fprepare_wave_worktrees.py` if\n   `.claude\u002F` is missing). Pass the PR head and wave JSON as argv to that\n   helper only — never interpolate PR head, task ids, or SHAs into a shell\n   git command. The helper validates the ref with `git check-ref-format\n   --normalize`, validates each id with `^TASK-[0-9]+$`, maps `\u002F` in the\n   branch name to a single `.worktrees\u002F` path segment, and calls\n   `git worktree add` \u002F `git branch` via `subprocess.run` list argv (no\n   `shell=True`). It reuses leftover dirs (`git reset --hard` to PR HEAD),\n   falls back to `\u002Ftmp\u002Fpr-resolve-\u003CTASK-ID>`, and creates the task branch\n   without moving PR HEAD when both add sites fail. Use the JSON\n   `worktree` \u002F `branch` fields; do not assemble those paths yourself.\n   Helper `add` JSON `isolated: true` means the checkout exists; it does not copy\n   `tasks_path` — the caller must copy it. After each successful\n   `git worktree add` (including the `\u002Ftmp` fallback) and after\n   reuse\u002Freset, copy the frozen gitignored\n   `tasks_path` from the PR worktree into that worktree (same filename)\n   so `test -f \u003Cworktree>\u002F\u003Ctasks_path>` is true and the bytes match the\n   PR-worktree manifest. A resolver restricted to the worktree must read\n   the assigned task without consulting the PR worktree. If copy is\n   skipped, pass an absolute PR-worktree `tasks_path` and say so in the\n   brief.\n3. If any wave task still has no isolated dir after reuse\u002Freset and both\n   add sites, do not stop the whole wave. Dispatch `issue_resolver` with\n   `resolve-next-task` for every task whose worktree succeeded. Leave ids\n   with no isolated dir `[open]` (undispatched). Sequential one-task mode\n   is not used solely because of leftovers.\n4. Otherwise — only when every wave task has an isolated checkout —\n   dispatch **all wave resolvers in one turn**, in parallel (same protocol\n   as panel). Issue the calls in a **single** response. One call per\n   response is a protocol failure.\n5. Each brief: `issue_resolver`, `resolve-next-task`, `tasks_path`,\n   `task_id`, worktree path, task branch, PR head branch name. Git cwd is\n   the worktree; `tasks_path` is the copied file in that worktree or the\n   absolute PR-worktree manifest path named in this brief. \"Do not push PR\n   head. Do not edit the tasks file or REVIEW_HISTORY.md.\"\n6. After all return: in TASK id order, run\n   `python3 ...\u002Fprepare_wave_worktrees.py cherry-pick --task-branch \u003Cbranch> [--] \u003Csha>`\n   for each resolver (the helper `git fetch`es the task branch, then\n   `git cherry-pick` with the SHA after `--`). Success per task id is\n   resolver `status=ok` plus a commit SHA that applies cleanly.\n   Missing SHA, blocked resolver, or status other than `ok`: skip\n   cherry-pick (no placeholder SHA); leave that id `[open]`. On any\n   non-zero cherry-pick (conflict, bad or missing object) the helper runs\n   `git cherry-pick --abort`; leave that id `[open]`, continue with the\n   next wave task. HEAD must not remain in a cherry-pick or merge state.\n   Do not push PR head. Undispatched fallback tasks stay `[open]`. Never\n   hand-build `git cherry-pick` \u002F `git fetch` lines from PR head or SHA\n   strings.\n7. After cherry-picks (success, conflict abort, or missing sha) and after\n   every wave (success, conflict, or dispatch failure), always run\n   `python3 ...\u002Fprepare_wave_worktrees.py prune --pr-head \u003Cref> \u003Cwave-json>`.\n   The helper runs `git worktree remove --force` on the wave paths,\n   `git worktree prune`, deletes local `\u003Cpr-head>-\u003CTASK-ID>` branches, and\n   removes leftover `\u002Ftmp\u002Fpr-resolve-\u003CTASK-ID>` dirs that are not\n   registered worktrees. Recreate them on the next attempt. On\n   abort-if-merged and on resolve-loop exit, prune any remaining\n   `.worktrees\u002F\u003Cpr-head>-*` paths and matching `\u002Ftmp` fallbacks\n   (worktree\u002Fbranch\u002Ftmp cleanup, not only child cancel\u002Farchive) before\n   emitting JSON. Never hand-build those git lines from PR head strings.\n8. After cherry-picks, stop. Never push `origin\u002F\u003Cpr-head>` in this skill.\n   Worktree git (add, cherry-pick, prune) is owned by `dispatch-resolve-wave`\n   via `prepare_wave_worktrees.py`. The orchestrator keeps mark-done,\n   log-progress, and is the only `git push` of `origin\u002F\u003Cpr-head>` — marks\n   `[done]` only the ids whose cherry-pick landed cleanly (not the whole wave),\n   follows `log-progress`, and is the only `git push` of `origin\u002F\u003Cpr-head>` —\n   and only after at least one clean pick and only if HEAD is not in a\n   cherry-pick or merge state.\n\n## Resolve-wave retry cap\n\nA wave with no successful cherry-pick, or a task that conflicted \u002F returned\n`blocked` \u002F produced no sha, is one failed attempt for those task ids.\nCap retries at **3** (short backoff), same as other failure classes.\nConflicted or blocked ids are not immediately eligible for another\nidentical parallel wave in the same run beyond this cap. After the cap,\nleave those tasks `[open]` and stop this skill; the orchestrator exits the\nresolve loop with `open_task_ids` and continues panel\u002Fverify\u002Frisk. Do not\ndispatch a fourth identical parallel wave for the same ids in this run.\n\n## Harness\n\n- Cursor: one `Task` call per wave task in a single message (`issue_resolver`).\n- Claude Code: one Agent call per wave task using `issue_resolver`.\n\n## Guardrails\n\n- Never push `origin\u002F\u003Cpr-head>` (orchestrator pushes after successful picks)\n- Never rebase\n- Never force-push\n- Never open a second PR\n- Never exceed the script's wave\n- Never implement fixes\n- Never interpolate PR head refs or commit SHAs into a shell git command;\n  run `prepare_wave_worktrees.py` instead\n",{"name":5,"metadata":24,"description":7},{"loadout.sha":25,"loadout.source":26,"loadout.managed":27},"49eb9bd","skills\u002Fdispatch-resolve-wave\u002FSKILL.md","true","main","f6497b6aafdead5bb7cf244894b930f699a53e46",".claude\u002Fskills\u002Fdispatch-resolve-wave",1,"https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout","https:\u002F\u002Fgithub.com\u002Fsazlin","https:\u002F\u002Fseanazlin.com","Principal Engineer & Engineering Leader | AI Infrastructure & Marketplace Systems | Fractional CTO for 0-to-1 Startups","2026-11-14 20:13:46",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add sazlin\u002Floadout --skill dispatch-resolve-wave",{"tool":43,"command":44},"bunx","bunx skills add sazlin\u002Floadout --skill dispatch-resolve-wave",{"tool":46,"command":47},"pnpm","pnpm dlx skills add sazlin\u002Floadout --skill dispatch-resolve-wave","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdispatch-resolve-wave\u002FSKILL.md",true,[51,57],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"scripts\u002Fpartition_waves.py","SCRIPT","脚本",3950,"https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fpartition_waves.py",{"path":58,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":59,"sourceHtmlUrl":60},"scripts\u002Fprepare_wave_worktrees.py",11555,"https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdispatch-resolve-wave\u002Fscripts\u002Fprepare_wave_worktrees.py",[62,69,76,83,90,97],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7A06GGYSGA1EA79WNMD8JHJ","demo","Demo skill","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fmini_loadout\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7A06BTSB17JVWZ52Z3DQDYY","writing-skills","Use when creating new skills, editing existing skills, or verifying skills work before deployment","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-skills\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7A069Y4ZJZBPP82QJ9KBHTV","writing-plans","Use when you have a spec or requirements for a multi-step task, before touching code","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-plans\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7A0676YNTT61WYZ7241T3RB","verification-before-completion","Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Fskills\u002Fverification-before-completion\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7A065AY047VH6JBN19WFAZ9","using-superpowers","Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Fskills\u002Fusing-superpowers\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7A062FBQ3VPWZT3BP8DWYKA","using-git-worktrees","Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback","https:\u002F\u002Fgithub.com\u002Fsazlin\u002Floadout\u002Fblob\u002Fmain\u002Fskills\u002Fusing-git-worktrees\u002FSKILL.md",[],[]]