[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M734ZGDNFQP84F3Q698YCGGM":3,"skill-same:01M734ZGDNFQP84F3Q698YCGGM":49},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":32,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":34,"install":35,"resources":48},"01M734ZGDNFQP84F3Q698YCGGM","check-history",null,"Check closed PR history for issue.","Morrison-Lab\u002Fai-config","Morrison-Lab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F104235520?v=4","01M734G0451S2G35MN4HFQK7XP","Morrison Lab","01M734G059BYFDFQA8YTPDGPMH",4,"MIT","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fcheck-history\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: check-history\ndescription: \"Check closed PR history for issue.\"\nuser-invocable: false\nallowed-tools:\n  - Bash\n  - Read\n---\n\n# check-history\n\nBefore implementing changes for an issue or MR, review the repository's merge\nhistory to understand what decisions were made previously and why.\n\n## When this fires (automatically)\n\n- Starting work on any issue that involves modifying existing code\n- Before proposing changes to CI\u002FCD templates, shared infrastructure, or\n  configuration files\n- When the change touches code that has been refactored in prior MRs\n\n## First: is the issue already done or in progress?\n\nBefore reviewing past history, confirm the issue still needs a *new* PR. Two\nchecks (skipping them wastes a whole issue-pick):\n\n- **Already resolved on `main`?** An issue can go stale — fixed incidentally by\n  a later change (e.g. a dependency bump). Re-read the issue body against current\n  `main` before implementing.\n\n  ```bash\n  gh issue view \u003CN> --json state,title,body | cat   # VIEW_ISSUE — also: is it already closed?\n  ```\n\n- **Existing open PR for it?** Search open PRs for one that already addresses the\n  issue, so you don't open a second, parallel PR for the same work.\n\n  ```bash\n  gh pr list --state open --json number,title,headRefName,body,author,assignees \\\n    --jq '.[] | select(((.body \u002F\u002F \"\") | test(\"#\u003CN>\\\\b\")) or (.title | test(\"#\u003CN>\\\\b\"))) | \"#\\(.number) \\(.title) [\\(.headRefName); \\(.author.login); assignees: \\([.assignees[].login] | join(\",\"))]\"'   # LIST_PRS\n  ```\n\n- **On a long-lived or foundational issue, the issue text AND any design-doc\n  status header can lag the code by several PRs.** A mature feature (a phased\n  effort, or anything with a dedicated design doc) may be partly or mostly\n  implemented even when the issue reads as unstarted. The work can land across\n  sibling PRs that never updated this issue or that header. Before scoping\n  *new* code, verify the **actual implementation state against the code**:\n  read the key source files implementing the feature and the test files, not just\n  the issue body. (Seen on sparta: issue #164 \u002F #240 read as \"phase 4b not\n  done\", but the core feature was already fully implemented and tested across\n  sibling PRs — nearly rebuilt already-completed work. The right move was to audit,\n  correct the stale issue\u002Fdoc status, and pick the genuine next slice.)\n\nIf an open PR already covers it and passes `memories\u002Freviewing-prs.md`'s\nscope test (opened by or assigned to the invoking user, explicitly requested,\nor authored by the GitHub Actions app), **drive that PR to clean** instead of\nre-implementing.\nScope establishes authorization, not that the branch is free: read the PR's\nclaim state first per [`claim-pr`](..\u002F..\u002Fshared\u002Fworkflow\u002Fclaim-pr.md), skip it\nwhile another session's claim is live (a `hold off` comment with a push or\ncomment in the last 2 hours), and take over an unclaimed or expired one by\nposting your own claim comment before the first push.\nIf it covers the issue but fails that test, leave both the PR and the issue\nalone and report that the PR exists, so the user can assign or name it.\nIf `main`\nalready satisfies the issue, stand it down and report — don't open a no-op PR.\nWhen the issue is only *partly* done, don't rebuild the done part: audit it,\ncorrect the stale issue\u002Fdoc status, and scope only the genuine remaining slice.\n\n## Procedure\n\n0. **Check the issue for an already-open PR.** Before writing any code, look at\n   the issue itself for an open PR that already addresses it (one whose body\n   says `Closes #\u003CN>`). A parallel session — or an issues-sweep run on another\n   machine — may already be on it; building anyway produces a duplicate PR.\n\n   **GitHub:**\n   ```bash\n   gh pr list --state open --limit 100 --json number,title,body,author,assignees \\\n     --jq '.[] | select((.body \u002F\u002F \"\") | test(\"(Closes|Fixes|Resolves) #\u003CN>\\\\b\"; \"i\")) | \"#\\(.number) \\(.title) [\\(.author.login); assignees: \\([.assignees[].login] | join(\",\"))]\"'   # LIST_PRS\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr list --state opened --per-page=50 --output json 2>\u002Fdev\u002Fnull \\\n     | jq -r '.[] | select((.description \u002F\u002F \"\") | test(\"(Closes|Fixes|Resolves) #\u003CN>\\\\b\"; \"i\")) | \"!\\(.iid) \\(.title) [\\(.author.username); assignees: \\([.assignees[].username] | join(\",\"))]\"'\n   ```\n\n   If an open PR already covers the issue, apply the scope test above to the\n   author and assignee logins the listing prints:\n   **review or extend it** when it passes and the same claim check clears\n   (another session's live claim still blocks you, and an expired one is taken\n   over by posting your own claim comment first), and leave it untouched\n   (reporting it to the user) when it fails.\n   Either way, do not open a competing one.\n   This catches *in-flight* work; the merged\u002Fclosed\n   history below catches *settled* decisions.\n\n1. **List recent merged MRs** touching the same area:\n\n   **GitHub:**\n   ```bash\n   gh pr list --state merged --limit 20 --json number,title,headRefName | cat   # LIST_PRS\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr list --merged --per-page=20 2>&1 | cat\n   ```\n\n2. **Identify relevant MRs** — look for titles\u002Fbranches that relate to the\n   files or subsystem you're about to modify.\n\n3. **Read their descriptions** for rationale:\n\n   **GitHub:**\n   ```bash\n   gh pr view \u003CN> --json body --jq '.body' | cat   # VIEW_PR\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr view \u003CN> 2>&1 | cat\n   ```\n\n4. **Check for contradictions** — if a prior MR\u002FPR explicitly chose approach A\n   and you're about to implement approach B, understand *why* A was chosen.\n   Common patterns:\n   - MR\u002FPR #X switched to strategy A for performance → don't blindly revert to B\n   - MR\u002FPR #Y removed code as dead → verify it's still dead before re-adding\n   - MR\u002FPR #Z added a guard\u002Fcheck → understand what it prevents before removing\n\n5. **Document the relationship** in your commit message or MR\u002FPR description:\n   - \"This resolves the incompatibility between #X and #Y\"\n   - \"Supersedes the approach from #Z because [reason]\"\n   - \"Preserves the intent of #X (no wasted resources) while also...\"\n\n## What to look for\n\n- **Intentional trade-offs** documented in MR descriptions\n- **Reverted MRs** — if something was tried and reverted, don't repeat it\n  without understanding why it failed\n- **Sequential MRs** that built on each other — your change might break an\n  assumption from a later MR\n- **Closed (not merged) MRs** — rejected approaches that shouldn't be retried\n\n## Output\n\nReport a brief summary to the user:\n- Which MRs are relevant to the current work\n- Any potential conflicts or trade-offs identified\n- Confirmation that the proposed approach is compatible with prior decisions\n\nIf a conflict is found, surface it before writing code — don't implement first\nand discover the regression later.\n\n## Proactive hook compliance\n\n- **`warn-stale-issue-edit.py`**: Intercepts `Write`, `Edit`, and `NotebookEdit` tool calls\n  and warns if an issue-driven modification begins without a fresh `VIEW_ISSUE` (`gh issue view \u003CN>`)\n  and remote default-branch check (`git fetch origin main`).\n  Running the initial inspection in this skill satisfies the guard proactively\n  (see [`memories\u002Fhooks.md`](..\u002F..\u002Fmemories\u002Fhooks.md)).\n","# check-history\n\nBefore implementing changes for an issue or MR, review the repository's merge\nhistory to understand what decisions were made previously and why.\n\n## When this fires (automatically)\n\n- Starting work on any issue that involves modifying existing code\n- Before proposing changes to CI\u002FCD templates, shared infrastructure, or\n  configuration files\n- When the change touches code that has been refactored in prior MRs\n\n## First: is the issue already done or in progress?\n\nBefore reviewing past history, confirm the issue still needs a *new* PR. Two\nchecks (skipping them wastes a whole issue-pick):\n\n- **Already resolved on `main`?** An issue can go stale — fixed incidentally by\n  a later change (e.g. a dependency bump). Re-read the issue body against current\n  `main` before implementing.\n\n  ```bash\n  gh issue view \u003CN> --json state,title,body | cat   # VIEW_ISSUE — also: is it already closed?\n  ```\n\n- **Existing open PR for it?** Search open PRs for one that already addresses the\n  issue, so you don't open a second, parallel PR for the same work.\n\n  ```bash\n  gh pr list --state open --json number,title,headRefName,body,author,assignees \\\n    --jq '.[] | select(((.body \u002F\u002F \"\") | test(\"#\u003CN>\\\\b\")) or (.title | test(\"#\u003CN>\\\\b\"))) | \"#\\(.number) \\(.title) [\\(.headRefName); \\(.author.login); assignees: \\([.assignees[].login] | join(\",\"))]\"'   # LIST_PRS\n  ```\n\n- **On a long-lived or foundational issue, the issue text AND any design-doc\n  status header can lag the code by several PRs.** A mature feature (a phased\n  effort, or anything with a dedicated design doc) may be partly or mostly\n  implemented even when the issue reads as unstarted. The work can land across\n  sibling PRs that never updated this issue or that header. Before scoping\n  *new* code, verify the **actual implementation state against the code**:\n  read the key source files implementing the feature and the test files, not just\n  the issue body. (Seen on sparta: issue #164 \u002F #240 read as \"phase 4b not\n  done\", but the core feature was already fully implemented and tested across\n  sibling PRs — nearly rebuilt already-completed work. The right move was to audit,\n  correct the stale issue\u002Fdoc status, and pick the genuine next slice.)\n\nIf an open PR already covers it and passes `memories\u002Freviewing-prs.md`'s\nscope test (opened by or assigned to the invoking user, explicitly requested,\nor authored by the GitHub Actions app), **drive that PR to clean** instead of\nre-implementing.\nScope establishes authorization, not that the branch is free: read the PR's\nclaim state first per [`claim-pr`](..\u002F..\u002Fshared\u002Fworkflow\u002Fclaim-pr.md), skip it\nwhile another session's claim is live (a `hold off` comment with a push or\ncomment in the last 2 hours), and take over an unclaimed or expired one by\nposting your own claim comment before the first push.\nIf it covers the issue but fails that test, leave both the PR and the issue\nalone and report that the PR exists, so the user can assign or name it.\nIf `main`\nalready satisfies the issue, stand it down and report — don't open a no-op PR.\nWhen the issue is only *partly* done, don't rebuild the done part: audit it,\ncorrect the stale issue\u002Fdoc status, and scope only the genuine remaining slice.\n\n## Procedure\n\n0. **Check the issue for an already-open PR.** Before writing any code, look at\n   the issue itself for an open PR that already addresses it (one whose body\n   says `Closes #\u003CN>`). A parallel session — or an issues-sweep run on another\n   machine — may already be on it; building anyway produces a duplicate PR.\n\n   **GitHub:**\n   ```bash\n   gh pr list --state open --limit 100 --json number,title,body,author,assignees \\\n     --jq '.[] | select((.body \u002F\u002F \"\") | test(\"(Closes|Fixes|Resolves) #\u003CN>\\\\b\"; \"i\")) | \"#\\(.number) \\(.title) [\\(.author.login); assignees: \\([.assignees[].login] | join(\",\"))]\"'   # LIST_PRS\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr list --state opened --per-page=50 --output json 2>\u002Fdev\u002Fnull \\\n     | jq -r '.[] | select((.description \u002F\u002F \"\") | test(\"(Closes|Fixes|Resolves) #\u003CN>\\\\b\"; \"i\")) | \"!\\(.iid) \\(.title) [\\(.author.username); assignees: \\([.assignees[].username] | join(\",\"))]\"'\n   ```\n\n   If an open PR already covers the issue, apply the scope test above to the\n   author and assignee logins the listing prints:\n   **review or extend it** when it passes and the same claim check clears\n   (another session's live claim still blocks you, and an expired one is taken\n   over by posting your own claim comment first), and leave it untouched\n   (reporting it to the user) when it fails.\n   Either way, do not open a competing one.\n   This catches *in-flight* work; the merged\u002Fclosed\n   history below catches *settled* decisions.\n\n1. **List recent merged MRs** touching the same area:\n\n   **GitHub:**\n   ```bash\n   gh pr list --state merged --limit 20 --json number,title,headRefName | cat   # LIST_PRS\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr list --merged --per-page=20 2>&1 | cat\n   ```\n\n2. **Identify relevant MRs** — look for titles\u002Fbranches that relate to the\n   files or subsystem you're about to modify.\n\n3. **Read their descriptions** for rationale:\n\n   **GitHub:**\n   ```bash\n   gh pr view \u003CN> --json body --jq '.body' | cat   # VIEW_PR\n   ```\n\n   **GitLab:**\n   ```bash\n   glab mr view \u003CN> 2>&1 | cat\n   ```\n\n4. **Check for contradictions** — if a prior MR\u002FPR explicitly chose approach A\n   and you're about to implement approach B, understand *why* A was chosen.\n   Common patterns:\n   - MR\u002FPR #X switched to strategy A for performance → don't blindly revert to B\n   - MR\u002FPR #Y removed code as dead → verify it's still dead before re-adding\n   - MR\u002FPR #Z added a guard\u002Fcheck → understand what it prevents before removing\n\n5. **Document the relationship** in your commit message or MR\u002FPR description:\n   - \"This resolves the incompatibility between #X and #Y\"\n   - \"Supersedes the approach from #Z because [reason]\"\n   - \"Preserves the intent of #X (no wasted resources) while also...\"\n\n## What to look for\n\n- **Intentional trade-offs** documented in MR descriptions\n- **Reverted MRs** — if something was tried and reverted, don't repeat it\n  without understanding why it failed\n- **Sequential MRs** that built on each other — your change might break an\n  assumption from a later MR\n- **Closed (not merged) MRs** — rejected approaches that shouldn't be retried\n\n## Output\n\nReport a brief summary to the user:\n- Which MRs are relevant to the current work\n- Any potential conflicts or trade-offs identified\n- Confirmation that the proposed approach is compatible with prior decisions\n\nIf a conflict is found, surface it before writing code — don't implement first\nand discover the regression later.\n\n## Proactive hook compliance\n\n- **`warn-stale-issue-edit.py`**: Intercepts `Write`, `Edit`, and `NotebookEdit` tool calls\n  and warns if an issue-driven modification begins without a fresh `VIEW_ISSUE` (`gh issue view \u003CN>`)\n  and remote default-branch check (`git fetch origin main`).\n  Running the initial inspection in this skill satisfies the guard proactively\n  (see [`memories\u002Fhooks.md`](..\u002F..\u002Fmemories\u002Fhooks.md)).\n",{"name":5,"description":7,"allowed-tools":24,"user-invocable":20},[25,26],"Bash","Read","main","0f720481f8ce3b3d9fef485f6e230d8b3035748e","skills\u002Fcheck-history",1,"https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config","https:\u002F\u002Fmorrison-lab.github.io\u002Fai-config\u002F","https:\u002F\u002Fgithub.com\u002FMorrison-Lab","2026-11-15 04:39:15",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add Morrison-Lab\u002Fai-config --skill check-history",{"tool":41,"command":42},"bunx","bunx skills add Morrison-Lab\u002Fai-config --skill check-history",{"tool":44,"command":45},"pnpm","pnpm dlx skills add Morrison-Lab\u002Fai-config --skill check-history","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fcheck-history\u002FSKILL.md",true,[],[50,57,64,71,78,85],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M735BVYCGR5VEC65AFDVSNKW","wrap-up","End-of-session verification and UMS.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwrap-up\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M735BTH469T1CQ8G2Q9GENCR","workaround-watcher","Watch upstream issue for fix.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fworkaround-watcher\u002FSKILL.md",[],[],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M735BRKZWFANSQ0GM380B4TD","wn","→ whats-next.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwn\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M735BPE9V830B3B6NGXQ56J6","whats-next","Recommend concrete next step or action.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwhats-next\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M735BMHQ60EAD5B7QAWXC5A0","wait-for-results","Wait for pipeline\u002FCI results.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwait-for-results\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M735BKGZHPN1PEMJ7YBD5VJ8","wai","→ send-notes.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwai\u002FSKILL.md",[],[]]