[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7350RF0KVESC91CEPPN05Q0":3,"skill-same:01M7350RF0KVESC91CEPPN05Q0":53},{"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":32,"commitSha":33,"skillDirectory":34,"forksCount":35,"repositoryHtmlUrl":36,"repositoryHomepageUrl":37,"ownerHtmlUrl":38,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":39,"install":40,"resources":52},"01M7350RF0KVESC91CEPPN05Q0","consolidate-memory",null,"Merge redundant memories.","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\u002Fconsolidate-memory\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: consolidate-memory\ndescription: \"Merge redundant memories.\"\nuser-invocable: true\nallowed-tools:\n  - Bash\n  - Read\n  - Write\n  - Edit\n  - Glob\n  - Grep\n---\n\n# consolidate-memory — merge redundant memory entries into one canonical\n\nThe memory corpus (the topical files indexed in `memories\u002FMEMORY.md`, plus\nper-repo entries in `~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`)\naccretes near-duplicates: two sessions each record the same lesson in different\nwords, or a fact lands in both a general file and a repo-specific one. This skill collapses a\ngenuine-duplicate cluster into **one canonical entry** — unioning the facts and\nrepointing any cross-links — so the corpus shrinks but no fact and no\n`[[link]]` breaks.\n\nIt is the memory-corpus complement of `consolidate-skills`: same shape, different\ndata model. The key difference is that **memories are not invoked by name**, so\nthere is no alias stub to leave behind — a redundant entry is *removed* after its\nunique content is folded into the survivor. The detection half — finding and\nclassifying the overlap — is delegated to `find-overlap`; this skill adds the\nmerge action on top.\n\n## When this fires\n\n- \"consolidate memory\", \"consolidate memories\", \"merge duplicate memories\",\n  \"dedupe memories\", \"collapse redundant memory entries\", `\u002Fconsolidate-memory`.\n- You notice two memory entries that assert the same fact in different words.\n- A `find-overlap` run over `memories\u002F` reports a genuine-duplicate cluster.\n\n## The one distinction that matters\n\nMost \"overlap\" is **not** a duplicate. Before merging anything, classify each\ncluster into exactly one of three buckets:\n\n1. **Intentional scope layering — LEAVE ALONE.** A general rule in\n   `preferences.md` *and* a concrete instance of it in a repo's project memory\n   (`~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`) is deliberate — the general file states the\n   principle, the repo file pins the specifics. Collapsing them loses either the\n   generality or the specifics. Tell: the entries live in different scope files\n   and one is the named application of the other.\n2. **Adjacent-but-distinct — LEAVE ALONE (maybe cross-link).** Same topic,\n   different facts (e.g. two `tools.md` bullets about the same tool but covering\n   different flags). Merging them buries one fact inside another. If they should\n   point at each other, add a `[[link]]` — don't merge.\n3. **Genuine duplicate — CONSOLIDATE.** Two or more entries, in the same scope,\n   that assert the **same fact** to the same end. This is the only bucket this\n   skill acts on.\n\nIf you can't articulate what fact would be *lost* by keeping only one, it's a\nduplicate. If you can, it isn't.\n\n## Procedure\n\n### 1. Detect overlap — delegate to `find-overlap`\n\nDon't re-implement the audit; run **`find-overlap`** with scope `memories\u002F`. It\ngathers every entry across the memory files, clusters by shared subject, reads\nthe surrounding text, and returns each cluster already sorted into the three\nbuckets above (with a recommended disposition). Work from its report.\n\nIf `find-overlap` is unavailable, fall back to its inline pass — list the\nheadings\u002Fbullets across `memories\u002F*.md` and per-repo project memories\n(`~\u002F.claude\u002Fprojects\u002F*\u002Fmemory\u002F`), cluster by subject, then **read the full\nsurrounding context of every member** before classifying.\n\n### 2. Keep only the genuine-duplicate clusters\n\nFrom `find-overlap`'s report, act **only** on bucket 3. Leave intentional scope\nlayering (bucket 1) untouched, and add a `[[link]]` to adjacent-but-distinct\nclusters (bucket 2) only if they want one — never a merge.\n\n### 3. Propose the consolidation plan — get approval first\n\nLike `consolidate-skills` and `heal-skill`, do not mutate before the user signs\noff. Present, per cluster:\n\n- **The members** (file + heading\u002Fbullet) and why they're genuine duplicates\n  (not scope layering, not adjacent-but-distinct).\n- **The chosen canonical home** — which scope file the merged entry belongs in.\n  A fact true everywhere belongs in the matching general file (one of the\n  topical files indexed in `memories\u002FMEMORY.md`); a repo-specific fact belongs in that repo's\n  Claude project memory (`~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`). State why.\n- **What gets folded in** — every non-obvious detail and *why*-clause unique to\n  an absorbed entry that must survive into the canonical.\n- **Which entries get removed** (all non-canonical members).\n- **Links to repoint** — any `[[link]]` pointing at an absorbed entry's topic\n  (see step 6).\n\n### 4. Build the canonical entry\n\nIn the chosen scope file:\n\n- **Union the facts** — fold every distinct detail from the absorbed entries\n  into one coherent entry. Keep the *why*, not just the *what*; don't just\n  staple the bullets together.\n- Match the file's existing entry shape (bullet vs `##` section, heading style,\n  terseness) so it reads like its neighbors.\n- Keep it concise — bullet points, not prose, per `record-learnings`.\n\n### 5. Remove the absorbed entries\n\nDelete the now-redundant copies. Unlike `consolidate-skills`, there is **no\nalias stub** — nothing invokes a memory by name, so a removed entry leaves no\ndangling invocation. Only the survivor remains. Never delete a whole memory\n*file* unless consolidation empties it.\n\n### 6. Fix every dangling reference\n\nA merged-away entry may be referenced by a `[[name]]` cross-link. List every\n`[[...]]` use across the memory files and skill bodies, then repoint any that\npointed at an absorbed entry at the surviving canonical:\n\n```bash\ngrep -rn \"\\[\\[\" memories\u002F skills\u002F*\u002FSKILL.md CLAUDE.md \\\n  ~\u002F.claude\u002Fprojects\u002F*\u002Fmemory\u002F 2>\u002Fdev\u002Fnull\n```\n\nIf you also named the absorbed entry in prose, grep that distinctive text and\nrepoint it too. Don't leave a reference resolving to an entry you removed.\n\n### 7. Validate, then ship via branch + PR\n\n```bash\npython3 scripts\u002Fvalidate-skills.py      # this skill's own frontmatter must pass\npython3 scripts\u002Fcheck-links.py          # no broken relative links\n```\n\nMemories live in the ai-config repo — never leave changes local-only. Branch +\nPR (not direct to main), request `the repository owner` as reviewer (`request-pr-review`),\nthen **ARDI to clean** (`ardi`).\n\n```bash\ngit checkout -b consolidate-memory-\u003Ctopic> origin\u002Fmain   # CREATE_BRANCH\n# build the canonical entry, remove the absorbed copies, repoint links\ngit add memories\u002F\u003Cfile>.md   # only the files you touched — never a bare\n                             # `git add memories\u002F` (sweeps in unrelated edits) or `git add -A`\ngit commit -m \"memories: consolidate \u003Ctopic> duplicates into one canonical entry\"   # COMMIT\n```\n\nPush as a separate Bash call, per [`check-before-pushing`](..\u002F..\u002Fshared\u002Fworkflow\u002Fcheck-before-pushing.md)'s \"Keep the commit in its own Bash call\":\n\n```bash\nrepo=\"${CLAUDE_PLUGIN_ROOT:-$(git -C ~\u002F.claude\u002Fskills\u002Fconsolidate-memory rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || pwd)}\"\ngit -C \"$repo\" push -u origin HEAD   # PUSH\n```\n\nOpen the PR in a separate Bash call:\n\n```bash\ngh pr create --fill   # CREATE_PR\n```\n\n## Relationship to other skills\n\n- **`find-overlap`** — the read-only detector this skill delegates its audit to;\n  it finds and classifies the overlapping memory clusters, consolidate-memory\n  merges the genuine duplicates. (find-overlap : consolidate-memory :: `pr-status`\n  : `ardi`.)\n- **`consolidate-skills`** — the same operation for the *skills* corpus. The\n  difference: skills preserve every invocation name via alias stubs;\n  consolidate-memory removes the redundant entry outright (memories aren't\n  invoked).\n- **`memorize` \u002F `record-learnings`** — they *write* memory entries; this *merges*\n  the redundant ones they accumulate. Step 3's \"check existing notes\" in those\n  skills prevents most duplicates; this is the cleanup when one slips through.\n- **`link-skills`** — the skills-corpus analogue: for adjacent-but-distinct\n  *skills* that should cross-reference each other but not merge. For memory\n  cross-links, add the `[[link]]` manually (as step 2 describes) — `link-skills`\n  only scans `skills\u002F*\u002FSKILL.md`.\n\n## Anti-patterns\n\n- ❌ Collapsing **intentional scope layering** (a general rule plus its\n  repo-specific instance) — that's deliberate, not a duplicate.\n- ❌ Merging **adjacent-but-distinct** entries and burying one fact inside\n  another.\n- ❌ Dropping a *why*-clause or a non-obvious detail on the floor — the canonical\n  must carry the **union** of the absorbed entries' content.\n- ❌ Leaving an absorbed entry's topic referenced by stale prose or a `[[link]]`.\n- ❌ Mutating before the user approves the plan.\n- ❌ `git add -A` \u002F a bare `git add` that sweeps in unrelated in-flight edits —\n  stage only the memory files you touched.\n","# consolidate-memory — merge redundant memory entries into one canonical\n\nThe memory corpus (the topical files indexed in `memories\u002FMEMORY.md`, plus\nper-repo entries in `~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`)\naccretes near-duplicates: two sessions each record the same lesson in different\nwords, or a fact lands in both a general file and a repo-specific one. This skill collapses a\ngenuine-duplicate cluster into **one canonical entry** — unioning the facts and\nrepointing any cross-links — so the corpus shrinks but no fact and no\n`[[link]]` breaks.\n\nIt is the memory-corpus complement of `consolidate-skills`: same shape, different\ndata model. The key difference is that **memories are not invoked by name**, so\nthere is no alias stub to leave behind — a redundant entry is *removed* after its\nunique content is folded into the survivor. The detection half — finding and\nclassifying the overlap — is delegated to `find-overlap`; this skill adds the\nmerge action on top.\n\n## When this fires\n\n- \"consolidate memory\", \"consolidate memories\", \"merge duplicate memories\",\n  \"dedupe memories\", \"collapse redundant memory entries\", `\u002Fconsolidate-memory`.\n- You notice two memory entries that assert the same fact in different words.\n- A `find-overlap` run over `memories\u002F` reports a genuine-duplicate cluster.\n\n## The one distinction that matters\n\nMost \"overlap\" is **not** a duplicate. Before merging anything, classify each\ncluster into exactly one of three buckets:\n\n1. **Intentional scope layering — LEAVE ALONE.** A general rule in\n   `preferences.md` *and* a concrete instance of it in a repo's project memory\n   (`~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`) is deliberate — the general file states the\n   principle, the repo file pins the specifics. Collapsing them loses either the\n   generality or the specifics. Tell: the entries live in different scope files\n   and one is the named application of the other.\n2. **Adjacent-but-distinct — LEAVE ALONE (maybe cross-link).** Same topic,\n   different facts (e.g. two `tools.md` bullets about the same tool but covering\n   different flags). Merging them buries one fact inside another. If they should\n   point at each other, add a `[[link]]` — don't merge.\n3. **Genuine duplicate — CONSOLIDATE.** Two or more entries, in the same scope,\n   that assert the **same fact** to the same end. This is the only bucket this\n   skill acts on.\n\nIf you can't articulate what fact would be *lost* by keeping only one, it's a\nduplicate. If you can, it isn't.\n\n## Procedure\n\n### 1. Detect overlap — delegate to `find-overlap`\n\nDon't re-implement the audit; run **`find-overlap`** with scope `memories\u002F`. It\ngathers every entry across the memory files, clusters by shared subject, reads\nthe surrounding text, and returns each cluster already sorted into the three\nbuckets above (with a recommended disposition). Work from its report.\n\nIf `find-overlap` is unavailable, fall back to its inline pass — list the\nheadings\u002Fbullets across `memories\u002F*.md` and per-repo project memories\n(`~\u002F.claude\u002Fprojects\u002F*\u002Fmemory\u002F`), cluster by subject, then **read the full\nsurrounding context of every member** before classifying.\n\n### 2. Keep only the genuine-duplicate clusters\n\nFrom `find-overlap`'s report, act **only** on bucket 3. Leave intentional scope\nlayering (bucket 1) untouched, and add a `[[link]]` to adjacent-but-distinct\nclusters (bucket 2) only if they want one — never a merge.\n\n### 3. Propose the consolidation plan — get approval first\n\nLike `consolidate-skills` and `heal-skill`, do not mutate before the user signs\noff. Present, per cluster:\n\n- **The members** (file + heading\u002Fbullet) and why they're genuine duplicates\n  (not scope layering, not adjacent-but-distinct).\n- **The chosen canonical home** — which scope file the merged entry belongs in.\n  A fact true everywhere belongs in the matching general file (one of the\n  topical files indexed in `memories\u002FMEMORY.md`); a repo-specific fact belongs in that repo's\n  Claude project memory (`~\u002F.claude\u002Fprojects\u002F\u003Cpath>\u002Fmemory\u002F`). State why.\n- **What gets folded in** — every non-obvious detail and *why*-clause unique to\n  an absorbed entry that must survive into the canonical.\n- **Which entries get removed** (all non-canonical members).\n- **Links to repoint** — any `[[link]]` pointing at an absorbed entry's topic\n  (see step 6).\n\n### 4. Build the canonical entry\n\nIn the chosen scope file:\n\n- **Union the facts** — fold every distinct detail from the absorbed entries\n  into one coherent entry. Keep the *why*, not just the *what*; don't just\n  staple the bullets together.\n- Match the file's existing entry shape (bullet vs `##` section, heading style,\n  terseness) so it reads like its neighbors.\n- Keep it concise — bullet points, not prose, per `record-learnings`.\n\n### 5. Remove the absorbed entries\n\nDelete the now-redundant copies. Unlike `consolidate-skills`, there is **no\nalias stub** — nothing invokes a memory by name, so a removed entry leaves no\ndangling invocation. Only the survivor remains. Never delete a whole memory\n*file* unless consolidation empties it.\n\n### 6. Fix every dangling reference\n\nA merged-away entry may be referenced by a `[[name]]` cross-link. List every\n`[[...]]` use across the memory files and skill bodies, then repoint any that\npointed at an absorbed entry at the surviving canonical:\n\n```bash\ngrep -rn \"\\[\\[\" memories\u002F skills\u002F*\u002FSKILL.md CLAUDE.md \\\n  ~\u002F.claude\u002Fprojects\u002F*\u002Fmemory\u002F 2>\u002Fdev\u002Fnull\n```\n\nIf you also named the absorbed entry in prose, grep that distinctive text and\nrepoint it too. Don't leave a reference resolving to an entry you removed.\n\n### 7. Validate, then ship via branch + PR\n\n```bash\npython3 scripts\u002Fvalidate-skills.py      # this skill's own frontmatter must pass\npython3 scripts\u002Fcheck-links.py          # no broken relative links\n```\n\nMemories live in the ai-config repo — never leave changes local-only. Branch +\nPR (not direct to main), request `the repository owner` as reviewer (`request-pr-review`),\nthen **ARDI to clean** (`ardi`).\n\n```bash\ngit checkout -b consolidate-memory-\u003Ctopic> origin\u002Fmain   # CREATE_BRANCH\n# build the canonical entry, remove the absorbed copies, repoint links\ngit add memories\u002F\u003Cfile>.md   # only the files you touched — never a bare\n                             # `git add memories\u002F` (sweeps in unrelated edits) or `git add -A`\ngit commit -m \"memories: consolidate \u003Ctopic> duplicates into one canonical entry\"   # COMMIT\n```\n\nPush as a separate Bash call, per [`check-before-pushing`](..\u002F..\u002Fshared\u002Fworkflow\u002Fcheck-before-pushing.md)'s \"Keep the commit in its own Bash call\":\n\n```bash\nrepo=\"${CLAUDE_PLUGIN_ROOT:-$(git -C ~\u002F.claude\u002Fskills\u002Fconsolidate-memory rev-parse --show-toplevel 2>\u002Fdev\u002Fnull || pwd)}\"\ngit -C \"$repo\" push -u origin HEAD   # PUSH\n```\n\nOpen the PR in a separate Bash call:\n\n```bash\ngh pr create --fill   # CREATE_PR\n```\n\n## Relationship to other skills\n\n- **`find-overlap`** — the read-only detector this skill delegates its audit to;\n  it finds and classifies the overlapping memory clusters, consolidate-memory\n  merges the genuine duplicates. (find-overlap : consolidate-memory :: `pr-status`\n  : `ardi`.)\n- **`consolidate-skills`** — the same operation for the *skills* corpus. The\n  difference: skills preserve every invocation name via alias stubs;\n  consolidate-memory removes the redundant entry outright (memories aren't\n  invoked).\n- **`memorize` \u002F `record-learnings`** — they *write* memory entries; this *merges*\n  the redundant ones they accumulate. Step 3's \"check existing notes\" in those\n  skills prevents most duplicates; this is the cleanup when one slips through.\n- **`link-skills`** — the skills-corpus analogue: for adjacent-but-distinct\n  *skills* that should cross-reference each other but not merge. For memory\n  cross-links, add the `[[link]]` manually (as step 2 describes) — `link-skills`\n  only scans `skills\u002F*\u002FSKILL.md`.\n\n## Anti-patterns\n\n- ❌ Collapsing **intentional scope layering** (a general rule plus its\n  repo-specific instance) — that's deliberate, not a duplicate.\n- ❌ Merging **adjacent-but-distinct** entries and burying one fact inside\n  another.\n- ❌ Dropping a *why*-clause or a non-obvious detail on the floor — the canonical\n  must carry the **union** of the absorbed entries' content.\n- ❌ Leaving an absorbed entry's topic referenced by stale prose or a `[[link]]`.\n- ❌ Mutating before the user approves the plan.\n- ❌ `git add -A` \u002F a bare `git add` that sweeps in unrelated in-flight edits —\n  stage only the memory files you touched.\n",{"name":5,"description":7,"allowed-tools":24,"user-invocable":31},[25,26,27,28,29,30],"Bash","Read","Write","Edit","Glob","Grep",true,"main","0f720481f8ce3b3d9fef485f6e230d8b3035748e","skills\u002Fconsolidate-memory",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":41,"prompt":51,"zipAvailable":31},[42,45,48],{"tool":43,"command":44},"npx","npx skills add Morrison-Lab\u002Fai-config --skill consolidate-memory",{"tool":46,"command":47},"bunx","bunx skills add Morrison-Lab\u002Fai-config --skill consolidate-memory",{"tool":49,"command":50},"pnpm","pnpm dlx skills add Morrison-Lab\u002Fai-config --skill consolidate-memory","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fconsolidate-memory\u002FSKILL.md",[],[54,61,68,75,82,89],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"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":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"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":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M735BRKZWFANSQ0GM380B4TD","wn","→ whats-next.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwn\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M735BKGZHPN1PEMJ7YBD5VJ8","wai","→ send-notes.","https:\u002F\u002Fgithub.com\u002FMorrison-Lab\u002Fai-config\u002Fblob\u002Fmain\u002Fskills\u002Fwai\u002FSKILL.md",[],[]]