[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-letta-letta-filesystem-to-memfs":3,"mdc--xbotqj-key":34,"related-org-letta-letta-filesystem-to-memfs":1673,"related-repo-letta-letta-filesystem-to-memfs":1836},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":29,"sourceUrl":32,"mdContent":33},"letta-filesystem-to-memfs","migrate files to MemFS","Migrates deprecated Letta Filesystem folders\u002Ffiles to MemFS using markdown document corpora, chunking, local lexical search, and QMD semantic search via the memfs-search skill. Use when replacing folders.files.upload, working with PDFs or document QA, or emulating open_file, grep_file, and search_file behavior.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"letta","Letta","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fletta.png","letta-ai",[13,17,20],{"name":14,"slug":15,"type":16},"Data Engineering","data-engineering","tag",{"name":18,"slug":19,"type":16},"Migration","migration",{"name":21,"slug":22,"type":16},"Search","search",127,"https:\u002F\u002Fgithub.com\u002Fletta-ai\u002Fskills","2026-07-13T06:25:52.385243","MIT",20,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],"A shared repository for skills. Intended to be used with Letta Code, Claude Code, Codex CLI, and other agents that support skills.","https:\u002F\u002Fgithub.com\u002Fletta-ai\u002Fskills\u002Ftree\u002FHEAD\u002Fletta\u002Fletta-filesystem-to-memfs","---\nname: letta-filesystem-to-memfs\ndescription: Migrates deprecated Letta Filesystem folders\u002Ffiles to MemFS using markdown document corpora, chunking, local lexical search, and QMD semantic search via the memfs-search skill. Use when replacing folders.files.upload, working with PDFs or document QA, or emulating open_file, grep_file, and search_file behavior.\nlicense: MIT\n---\n\n# Letta Filesystem to MemFS\n\nUse this skill when a user wants the old Letta Filesystem behavior: upload documents, chunk them, attach them to an agent, and let the agent search\u002Fopen them.\n\nMemFS is not the same product. It is git-backed markdown memory. To mimic the old workflow, use the bundled CLI:\n\n1. Extract PDFs\u002Fdocs to markdown text.\n2. Chunk the text into stable markdown files under `documents\u002F\u003Ccorpus>\u002F\u003Cdoc>\u002Fchunks\u002F`.\n3. Write a small pinned index under `system\u002Ffilesystem\u002F\u003Ccorpus>.md`.\n4. Index only the corpus chunk files in QMD for semantic search.\n5. Review the MemFS git diff. Commit only if asked.\n\n## Quick workflow\n\n```bash\n# Set this to the skill directory shown in the skill load header.\nSKILL_DIR=\"\u002Fpath\u002Fto\u002Fletta-filesystem-to-memfs\"\n\n# From any directory. MEMORY_DIR should point at the target agent's memfs repo.\nuv run --with pymupdf \\\n  \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" ingest \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  --source .\u002Fdocs\u002F \\\n  --source .\u002Fguide.pdf \\\n  --source https:\u002F\u002Farxiv.org\u002Fpdf\u002F2310.08560\n\ncd \"$MEMORY_DIR\"\ngit status --short\ngit diff --stat\n```\n\nSearch the chunk corpus lexically:\n\n```bash\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" search \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  \"memory hierarchy\" \\\n  -n 5\n```\n\nSemantic search with QMD. The CLI creates a corpus-scoped QMD collection over chunk files only:\n\n```bash\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd setup \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs\n\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd query \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  \"memory hierarchy\" \\\n  -n 5\n```\n\nUse `qmd reindex` after changing corpus files, and `qmd search` \u002F `qmd vsearch` when you specifically want keyword-only or vector-only search.\n\n## Layout\n\nThe ingest script writes:\n\n```txt\nsystem\u002Ffilesystem\u002F\u003Ccorpus>.md\n  Pinned index and operating instructions for the corpus.\n\ndocuments\u002F\u003Ccorpus>\u002Fmanifest.md\n  Corpus manifest.\n\ndocuments\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fmanifest.md\n  Per-document metadata and chunk table.\n\ndocuments\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fchunks\u002Fchunk-0001.md\n  Chunk content with frontmatter metadata.\n\ndocuments\u002F\u003Ccorpus>\u002Fchunks.jsonl\n  Machine-readable chunk export for custom indexing or debugging.\n```\n\n## Old API mapping\n\n| Old Filesystem | MemFS mimic |\n|---|---|\n| `folders.create` | `--corpus \u003Cname>` creates `documents\u002F\u003Ccorpus>\u002F` |\n| `folders.files.upload` | `letta_fs_to_memfs.py ingest --source \u003Cfile-or-directory-or-url>` |\n| OCR\u002Fchunk\u002Fembed job | Extract + chunk locally; `qmd setup` \u002F `qmd reindex` for semantic search |\n| `agents.folders.attach` | Enable MemFS, then review and sync repo changes when appropriate |\n| `open_file` | Read chunk markdown files by path |\n| `grep_file` | `rg` or `letta_fs_to_memfs.py search` |\n| `search_file` | `letta_fs_to_memfs.py qmd query` over the corpus chunk collection |\n\n## Heuristics\n\n- Use `system\u002Ffilesystem\u002F\u003Ccorpus>.md` for the small always-visible index only.\n- Do not pin full documents into `system\u002F`; it will bloat the prompt.\n- Keep chunks outside `system\u002F`, usually under `documents\u002F\u003Ccorpus>\u002F...`.\n- Passing a directory to `--source` recursively ingests supported files (`.pdf`, `.md`, `.txt`, `.json`, `.csv`, `.html`, `.xml`).\n- Use `--glob` \u002F `--exclude` for messy directories. Defaults exclude `.git`, `node_modules`, `.venv`, and `__pycache__`.\n- URL downloads default to `--max-download-mb 100`; set `0` for unlimited.\n- Re-ingesting the same document slug replaces that document's old chunk directory, so stale chunks do not survive chunk-size changes.\n- Use chunk sizes around 2,000-4,000 characters with 200-500 character overlap.\n- Use the CLI's `qmd` subcommands when the user needs semantic search over many chunks.\n- Preserve source URLs, checksums, page markers, chunk numbers, and corpus names in the generated files.\n\n## CLI reference\n\n```bash\nuv run --with pymupdf \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" ingest --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" search --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd setup --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd query --help\n```\n\nCompatibility wrappers remain for older examples:\n\n```bash\nuv run --with pymupdf \"$SKILL_DIR\u002Fscripts\u002Fingest_documents.py\" --memory-dir \"$MEMORY_DIR\" --corpus docs --source .\u002Fdocs\nuv run \"$SKILL_DIR\u002Fscripts\u002Fsearch_corpus.py\" --memory-dir \"$MEMORY_DIR\" --corpus docs --query \"refund policy\"\n```\n\n## PDF notes\n\nThe ingest script uses PyMuPDF when it sees a PDF. Run it with `uv run --with pymupdf ...`.\n\nFor scanned PDFs or complex tables, use the `tools\u002Fextracting-pdf-text` skill first, then ingest the extracted markdown\u002Ftext file with this skill.\n\nSee `references\u002Fdesign.md` for design notes and edge cases.\n",{"data":35,"body":36},{"name":4,"description":6,"license":26},{"type":37,"children":38},"root",[39,47,53,58,104,111,407,412,527,532,730,759,765,770,887,893,1084,1090,1298,1304,1468,1473,1623,1629,1641,1654,1667],{"type":40,"tag":41,"props":42,"children":43},"element","h1",{"id":4},[44],{"type":45,"value":46},"text","Letta Filesystem to MemFS",{"type":40,"tag":48,"props":49,"children":50},"p",{},[51],{"type":45,"value":52},"Use this skill when a user wants the old Letta Filesystem behavior: upload documents, chunk them, attach them to an agent, and let the agent search\u002Fopen them.",{"type":40,"tag":48,"props":54,"children":55},{},[56],{"type":45,"value":57},"MemFS is not the same product. It is git-backed markdown memory. To mimic the old workflow, use the bundled CLI:",{"type":40,"tag":59,"props":60,"children":61},"ol",{},[62,68,82,94,99],{"type":40,"tag":63,"props":64,"children":65},"li",{},[66],{"type":45,"value":67},"Extract PDFs\u002Fdocs to markdown text.",{"type":40,"tag":63,"props":69,"children":70},{},[71,73,80],{"type":45,"value":72},"Chunk the text into stable markdown files under ",{"type":40,"tag":74,"props":75,"children":77},"code",{"className":76},[],[78],{"type":45,"value":79},"documents\u002F\u003Ccorpus>\u002F\u003Cdoc>\u002Fchunks\u002F",{"type":45,"value":81},".",{"type":40,"tag":63,"props":83,"children":84},{},[85,87,93],{"type":45,"value":86},"Write a small pinned index under ",{"type":40,"tag":74,"props":88,"children":90},{"className":89},[],[91],{"type":45,"value":92},"system\u002Ffilesystem\u002F\u003Ccorpus>.md",{"type":45,"value":81},{"type":40,"tag":63,"props":95,"children":96},{},[97],{"type":45,"value":98},"Index only the corpus chunk files in QMD for semantic search.",{"type":40,"tag":63,"props":100,"children":101},{},[102],{"type":45,"value":103},"Review the MemFS git diff. Commit only if asked.",{"type":40,"tag":105,"props":106,"children":108},"h2",{"id":107},"quick-workflow",[109],{"type":45,"value":110},"Quick workflow",{"type":40,"tag":112,"props":113,"children":118},"pre",{"className":114,"code":115,"language":116,"meta":117,"style":117},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","# Set this to the skill directory shown in the skill load header.\nSKILL_DIR=\"\u002Fpath\u002Fto\u002Fletta-filesystem-to-memfs\"\n\n# From any directory. MEMORY_DIR should point at the target agent's memfs repo.\nuv run --with pymupdf \\\n  \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" ingest \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  --source .\u002Fdocs\u002F \\\n  --source .\u002Fguide.pdf \\\n  --source https:\u002F\u002Farxiv.org\u002Fpdf\u002F2310.08560\n\ncd \"$MEMORY_DIR\"\ngit status --short\ngit diff --stat\n","bash","",[119],{"type":40,"tag":74,"props":120,"children":121},{"__ignoreMap":117},[122,134,166,176,185,215,247,274,292,310,327,340,348,370,389],{"type":40,"tag":123,"props":124,"children":127},"span",{"class":125,"line":126},"line",1,[128],{"type":40,"tag":123,"props":129,"children":131},{"style":130},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[132],{"type":45,"value":133},"# Set this to the skill directory shown in the skill load header.\n",{"type":40,"tag":123,"props":135,"children":137},{"class":125,"line":136},2,[138,144,150,155,161],{"type":40,"tag":123,"props":139,"children":141},{"style":140},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[142],{"type":45,"value":143},"SKILL_DIR",{"type":40,"tag":123,"props":145,"children":147},{"style":146},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[148],{"type":45,"value":149},"=",{"type":40,"tag":123,"props":151,"children":152},{"style":146},[153],{"type":45,"value":154},"\"",{"type":40,"tag":123,"props":156,"children":158},{"style":157},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[159],{"type":45,"value":160},"\u002Fpath\u002Fto\u002Fletta-filesystem-to-memfs",{"type":40,"tag":123,"props":162,"children":163},{"style":146},[164],{"type":45,"value":165},"\"\n",{"type":40,"tag":123,"props":167,"children":169},{"class":125,"line":168},3,[170],{"type":40,"tag":123,"props":171,"children":173},{"emptyLinePlaceholder":172},true,[174],{"type":45,"value":175},"\n",{"type":40,"tag":123,"props":177,"children":179},{"class":125,"line":178},4,[180],{"type":40,"tag":123,"props":181,"children":182},{"style":130},[183],{"type":45,"value":184},"# From any directory. MEMORY_DIR should point at the target agent's memfs repo.\n",{"type":40,"tag":123,"props":186,"children":188},{"class":125,"line":187},5,[189,195,200,205,210],{"type":40,"tag":123,"props":190,"children":192},{"style":191},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[193],{"type":45,"value":194},"uv",{"type":40,"tag":123,"props":196,"children":197},{"style":157},[198],{"type":45,"value":199}," run",{"type":40,"tag":123,"props":201,"children":202},{"style":157},[203],{"type":45,"value":204}," --with",{"type":40,"tag":123,"props":206,"children":207},{"style":157},[208],{"type":45,"value":209}," pymupdf",{"type":40,"tag":123,"props":211,"children":212},{"style":140},[213],{"type":45,"value":214}," \\\n",{"type":40,"tag":123,"props":216,"children":218},{"class":125,"line":217},6,[219,224,229,234,238,243],{"type":40,"tag":123,"props":220,"children":221},{"style":146},[222],{"type":45,"value":223},"  \"",{"type":40,"tag":123,"props":225,"children":226},{"style":140},[227],{"type":45,"value":228},"$SKILL_DIR",{"type":40,"tag":123,"props":230,"children":231},{"style":157},[232],{"type":45,"value":233},"\u002Fscripts\u002Fletta_fs_to_memfs.py",{"type":40,"tag":123,"props":235,"children":236},{"style":146},[237],{"type":45,"value":154},{"type":40,"tag":123,"props":239,"children":240},{"style":157},[241],{"type":45,"value":242}," ingest",{"type":40,"tag":123,"props":244,"children":245},{"style":140},[246],{"type":45,"value":214},{"type":40,"tag":123,"props":248,"children":250},{"class":125,"line":249},7,[251,256,261,266,270],{"type":40,"tag":123,"props":252,"children":253},{"style":157},[254],{"type":45,"value":255},"  --memory-dir",{"type":40,"tag":123,"props":257,"children":258},{"style":146},[259],{"type":45,"value":260}," \"",{"type":40,"tag":123,"props":262,"children":263},{"style":140},[264],{"type":45,"value":265},"$MEMORY_DIR",{"type":40,"tag":123,"props":267,"children":268},{"style":146},[269],{"type":45,"value":154},{"type":40,"tag":123,"props":271,"children":272},{"style":140},[273],{"type":45,"value":214},{"type":40,"tag":123,"props":275,"children":277},{"class":125,"line":276},8,[278,283,288],{"type":40,"tag":123,"props":279,"children":280},{"style":157},[281],{"type":45,"value":282},"  --corpus",{"type":40,"tag":123,"props":284,"children":285},{"style":157},[286],{"type":45,"value":287}," product-docs",{"type":40,"tag":123,"props":289,"children":290},{"style":140},[291],{"type":45,"value":214},{"type":40,"tag":123,"props":293,"children":295},{"class":125,"line":294},9,[296,301,306],{"type":40,"tag":123,"props":297,"children":298},{"style":157},[299],{"type":45,"value":300},"  --source",{"type":40,"tag":123,"props":302,"children":303},{"style":157},[304],{"type":45,"value":305}," .\u002Fdocs\u002F",{"type":40,"tag":123,"props":307,"children":308},{"style":140},[309],{"type":45,"value":214},{"type":40,"tag":123,"props":311,"children":313},{"class":125,"line":312},10,[314,318,323],{"type":40,"tag":123,"props":315,"children":316},{"style":157},[317],{"type":45,"value":300},{"type":40,"tag":123,"props":319,"children":320},{"style":157},[321],{"type":45,"value":322}," .\u002Fguide.pdf",{"type":40,"tag":123,"props":324,"children":325},{"style":140},[326],{"type":45,"value":214},{"type":40,"tag":123,"props":328,"children":330},{"class":125,"line":329},11,[331,335],{"type":40,"tag":123,"props":332,"children":333},{"style":157},[334],{"type":45,"value":300},{"type":40,"tag":123,"props":336,"children":337},{"style":157},[338],{"type":45,"value":339}," https:\u002F\u002Farxiv.org\u002Fpdf\u002F2310.08560\n",{"type":40,"tag":123,"props":341,"children":343},{"class":125,"line":342},12,[344],{"type":40,"tag":123,"props":345,"children":346},{"emptyLinePlaceholder":172},[347],{"type":45,"value":175},{"type":40,"tag":123,"props":349,"children":351},{"class":125,"line":350},13,[352,358,362,366],{"type":40,"tag":123,"props":353,"children":355},{"style":354},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[356],{"type":45,"value":357},"cd",{"type":40,"tag":123,"props":359,"children":360},{"style":146},[361],{"type":45,"value":260},{"type":40,"tag":123,"props":363,"children":364},{"style":140},[365],{"type":45,"value":265},{"type":40,"tag":123,"props":367,"children":368},{"style":146},[369],{"type":45,"value":165},{"type":40,"tag":123,"props":371,"children":373},{"class":125,"line":372},14,[374,379,384],{"type":40,"tag":123,"props":375,"children":376},{"style":191},[377],{"type":45,"value":378},"git",{"type":40,"tag":123,"props":380,"children":381},{"style":157},[382],{"type":45,"value":383}," status",{"type":40,"tag":123,"props":385,"children":386},{"style":157},[387],{"type":45,"value":388}," --short\n",{"type":40,"tag":123,"props":390,"children":392},{"class":125,"line":391},15,[393,397,402],{"type":40,"tag":123,"props":394,"children":395},{"style":191},[396],{"type":45,"value":378},{"type":40,"tag":123,"props":398,"children":399},{"style":157},[400],{"type":45,"value":401}," diff",{"type":40,"tag":123,"props":403,"children":404},{"style":157},[405],{"type":45,"value":406}," --stat\n",{"type":40,"tag":48,"props":408,"children":409},{},[410],{"type":45,"value":411},"Search the chunk corpus lexically:",{"type":40,"tag":112,"props":413,"children":415},{"className":114,"code":414,"language":116,"meta":117,"style":117},"uv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" search \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  \"memory hierarchy\" \\\n  -n 5\n",[416],{"type":40,"tag":74,"props":417,"children":418},{"__ignoreMap":117},[419,455,478,493,513],{"type":40,"tag":123,"props":420,"children":421},{"class":125,"line":126},[422,426,430,434,438,442,446,451],{"type":40,"tag":123,"props":423,"children":424},{"style":191},[425],{"type":45,"value":194},{"type":40,"tag":123,"props":427,"children":428},{"style":157},[429],{"type":45,"value":199},{"type":40,"tag":123,"props":431,"children":432},{"style":146},[433],{"type":45,"value":260},{"type":40,"tag":123,"props":435,"children":436},{"style":140},[437],{"type":45,"value":228},{"type":40,"tag":123,"props":439,"children":440},{"style":157},[441],{"type":45,"value":233},{"type":40,"tag":123,"props":443,"children":444},{"style":146},[445],{"type":45,"value":154},{"type":40,"tag":123,"props":447,"children":448},{"style":157},[449],{"type":45,"value":450}," search",{"type":40,"tag":123,"props":452,"children":453},{"style":140},[454],{"type":45,"value":214},{"type":40,"tag":123,"props":456,"children":457},{"class":125,"line":136},[458,462,466,470,474],{"type":40,"tag":123,"props":459,"children":460},{"style":157},[461],{"type":45,"value":255},{"type":40,"tag":123,"props":463,"children":464},{"style":146},[465],{"type":45,"value":260},{"type":40,"tag":123,"props":467,"children":468},{"style":140},[469],{"type":45,"value":265},{"type":40,"tag":123,"props":471,"children":472},{"style":146},[473],{"type":45,"value":154},{"type":40,"tag":123,"props":475,"children":476},{"style":140},[477],{"type":45,"value":214},{"type":40,"tag":123,"props":479,"children":480},{"class":125,"line":168},[481,485,489],{"type":40,"tag":123,"props":482,"children":483},{"style":157},[484],{"type":45,"value":282},{"type":40,"tag":123,"props":486,"children":487},{"style":157},[488],{"type":45,"value":287},{"type":40,"tag":123,"props":490,"children":491},{"style":140},[492],{"type":45,"value":214},{"type":40,"tag":123,"props":494,"children":495},{"class":125,"line":178},[496,500,505,509],{"type":40,"tag":123,"props":497,"children":498},{"style":146},[499],{"type":45,"value":223},{"type":40,"tag":123,"props":501,"children":502},{"style":157},[503],{"type":45,"value":504},"memory hierarchy",{"type":40,"tag":123,"props":506,"children":507},{"style":146},[508],{"type":45,"value":154},{"type":40,"tag":123,"props":510,"children":511},{"style":140},[512],{"type":45,"value":214},{"type":40,"tag":123,"props":514,"children":515},{"class":125,"line":187},[516,521],{"type":40,"tag":123,"props":517,"children":518},{"style":157},[519],{"type":45,"value":520},"  -n",{"type":40,"tag":123,"props":522,"children":524},{"style":523},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[525],{"type":45,"value":526}," 5\n",{"type":40,"tag":48,"props":528,"children":529},{},[530],{"type":45,"value":531},"Semantic search with QMD. The CLI creates a corpus-scoped QMD collection over chunk files only:",{"type":40,"tag":112,"props":533,"children":535},{"className":114,"code":534,"language":116,"meta":117,"style":117},"uv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd setup \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs\n\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd query \\\n  --memory-dir \"$MEMORY_DIR\" \\\n  --corpus product-docs \\\n  \"memory hierarchy\" \\\n  -n 5\n",[536],{"type":40,"tag":74,"props":537,"children":538},{"__ignoreMap":117},[539,580,603,615,622,662,685,700,719],{"type":40,"tag":123,"props":540,"children":541},{"class":125,"line":126},[542,546,550,554,558,562,566,571,576],{"type":40,"tag":123,"props":543,"children":544},{"style":191},[545],{"type":45,"value":194},{"type":40,"tag":123,"props":547,"children":548},{"style":157},[549],{"type":45,"value":199},{"type":40,"tag":123,"props":551,"children":552},{"style":146},[553],{"type":45,"value":260},{"type":40,"tag":123,"props":555,"children":556},{"style":140},[557],{"type":45,"value":228},{"type":40,"tag":123,"props":559,"children":560},{"style":157},[561],{"type":45,"value":233},{"type":40,"tag":123,"props":563,"children":564},{"style":146},[565],{"type":45,"value":154},{"type":40,"tag":123,"props":567,"children":568},{"style":157},[569],{"type":45,"value":570}," qmd",{"type":40,"tag":123,"props":572,"children":573},{"style":157},[574],{"type":45,"value":575}," setup",{"type":40,"tag":123,"props":577,"children":578},{"style":140},[579],{"type":45,"value":214},{"type":40,"tag":123,"props":581,"children":582},{"class":125,"line":136},[583,587,591,595,599],{"type":40,"tag":123,"props":584,"children":585},{"style":157},[586],{"type":45,"value":255},{"type":40,"tag":123,"props":588,"children":589},{"style":146},[590],{"type":45,"value":260},{"type":40,"tag":123,"props":592,"children":593},{"style":140},[594],{"type":45,"value":265},{"type":40,"tag":123,"props":596,"children":597},{"style":146},[598],{"type":45,"value":154},{"type":40,"tag":123,"props":600,"children":601},{"style":140},[602],{"type":45,"value":214},{"type":40,"tag":123,"props":604,"children":605},{"class":125,"line":168},[606,610],{"type":40,"tag":123,"props":607,"children":608},{"style":157},[609],{"type":45,"value":282},{"type":40,"tag":123,"props":611,"children":612},{"style":157},[613],{"type":45,"value":614}," product-docs\n",{"type":40,"tag":123,"props":616,"children":617},{"class":125,"line":178},[618],{"type":40,"tag":123,"props":619,"children":620},{"emptyLinePlaceholder":172},[621],{"type":45,"value":175},{"type":40,"tag":123,"props":623,"children":624},{"class":125,"line":187},[625,629,633,637,641,645,649,653,658],{"type":40,"tag":123,"props":626,"children":627},{"style":191},[628],{"type":45,"value":194},{"type":40,"tag":123,"props":630,"children":631},{"style":157},[632],{"type":45,"value":199},{"type":40,"tag":123,"props":634,"children":635},{"style":146},[636],{"type":45,"value":260},{"type":40,"tag":123,"props":638,"children":639},{"style":140},[640],{"type":45,"value":228},{"type":40,"tag":123,"props":642,"children":643},{"style":157},[644],{"type":45,"value":233},{"type":40,"tag":123,"props":646,"children":647},{"style":146},[648],{"type":45,"value":154},{"type":40,"tag":123,"props":650,"children":651},{"style":157},[652],{"type":45,"value":570},{"type":40,"tag":123,"props":654,"children":655},{"style":157},[656],{"type":45,"value":657}," query",{"type":40,"tag":123,"props":659,"children":660},{"style":140},[661],{"type":45,"value":214},{"type":40,"tag":123,"props":663,"children":664},{"class":125,"line":217},[665,669,673,677,681],{"type":40,"tag":123,"props":666,"children":667},{"style":157},[668],{"type":45,"value":255},{"type":40,"tag":123,"props":670,"children":671},{"style":146},[672],{"type":45,"value":260},{"type":40,"tag":123,"props":674,"children":675},{"style":140},[676],{"type":45,"value":265},{"type":40,"tag":123,"props":678,"children":679},{"style":146},[680],{"type":45,"value":154},{"type":40,"tag":123,"props":682,"children":683},{"style":140},[684],{"type":45,"value":214},{"type":40,"tag":123,"props":686,"children":687},{"class":125,"line":249},[688,692,696],{"type":40,"tag":123,"props":689,"children":690},{"style":157},[691],{"type":45,"value":282},{"type":40,"tag":123,"props":693,"children":694},{"style":157},[695],{"type":45,"value":287},{"type":40,"tag":123,"props":697,"children":698},{"style":140},[699],{"type":45,"value":214},{"type":40,"tag":123,"props":701,"children":702},{"class":125,"line":276},[703,707,711,715],{"type":40,"tag":123,"props":704,"children":705},{"style":146},[706],{"type":45,"value":223},{"type":40,"tag":123,"props":708,"children":709},{"style":157},[710],{"type":45,"value":504},{"type":40,"tag":123,"props":712,"children":713},{"style":146},[714],{"type":45,"value":154},{"type":40,"tag":123,"props":716,"children":717},{"style":140},[718],{"type":45,"value":214},{"type":40,"tag":123,"props":720,"children":721},{"class":125,"line":294},[722,726],{"type":40,"tag":123,"props":723,"children":724},{"style":157},[725],{"type":45,"value":520},{"type":40,"tag":123,"props":727,"children":728},{"style":523},[729],{"type":45,"value":526},{"type":40,"tag":48,"props":731,"children":732},{},[733,735,741,743,749,751,757],{"type":45,"value":734},"Use ",{"type":40,"tag":74,"props":736,"children":738},{"className":737},[],[739],{"type":45,"value":740},"qmd reindex",{"type":45,"value":742}," after changing corpus files, and ",{"type":40,"tag":74,"props":744,"children":746},{"className":745},[],[747],{"type":45,"value":748},"qmd search",{"type":45,"value":750}," \u002F ",{"type":40,"tag":74,"props":752,"children":754},{"className":753},[],[755],{"type":45,"value":756},"qmd vsearch",{"type":45,"value":758}," when you specifically want keyword-only or vector-only search.",{"type":40,"tag":105,"props":760,"children":762},{"id":761},"layout",[763],{"type":45,"value":764},"Layout",{"type":40,"tag":48,"props":766,"children":767},{},[768],{"type":45,"value":769},"The ingest script writes:",{"type":40,"tag":112,"props":771,"children":775},{"className":772,"code":773,"language":774,"meta":117,"style":117},"language-txt shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","system\u002Ffilesystem\u002F\u003Ccorpus>.md\n  Pinned index and operating instructions for the corpus.\n\ndocuments\u002F\u003Ccorpus>\u002Fmanifest.md\n  Corpus manifest.\n\ndocuments\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fmanifest.md\n  Per-document metadata and chunk table.\n\ndocuments\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fchunks\u002Fchunk-0001.md\n  Chunk content with frontmatter metadata.\n\ndocuments\u002F\u003Ccorpus>\u002Fchunks.jsonl\n  Machine-readable chunk export for custom indexing or debugging.\n","txt",[776],{"type":40,"tag":74,"props":777,"children":778},{"__ignoreMap":117},[779,787,795,802,810,818,825,833,841,848,856,864,871,879],{"type":40,"tag":123,"props":780,"children":781},{"class":125,"line":126},[782],{"type":40,"tag":123,"props":783,"children":784},{},[785],{"type":45,"value":786},"system\u002Ffilesystem\u002F\u003Ccorpus>.md\n",{"type":40,"tag":123,"props":788,"children":789},{"class":125,"line":136},[790],{"type":40,"tag":123,"props":791,"children":792},{},[793],{"type":45,"value":794},"  Pinned index and operating instructions for the corpus.\n",{"type":40,"tag":123,"props":796,"children":797},{"class":125,"line":168},[798],{"type":40,"tag":123,"props":799,"children":800},{"emptyLinePlaceholder":172},[801],{"type":45,"value":175},{"type":40,"tag":123,"props":803,"children":804},{"class":125,"line":178},[805],{"type":40,"tag":123,"props":806,"children":807},{},[808],{"type":45,"value":809},"documents\u002F\u003Ccorpus>\u002Fmanifest.md\n",{"type":40,"tag":123,"props":811,"children":812},{"class":125,"line":187},[813],{"type":40,"tag":123,"props":814,"children":815},{},[816],{"type":45,"value":817},"  Corpus manifest.\n",{"type":40,"tag":123,"props":819,"children":820},{"class":125,"line":217},[821],{"type":40,"tag":123,"props":822,"children":823},{"emptyLinePlaceholder":172},[824],{"type":45,"value":175},{"type":40,"tag":123,"props":826,"children":827},{"class":125,"line":249},[828],{"type":40,"tag":123,"props":829,"children":830},{},[831],{"type":45,"value":832},"documents\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fmanifest.md\n",{"type":40,"tag":123,"props":834,"children":835},{"class":125,"line":276},[836],{"type":40,"tag":123,"props":837,"children":838},{},[839],{"type":45,"value":840},"  Per-document metadata and chunk table.\n",{"type":40,"tag":123,"props":842,"children":843},{"class":125,"line":294},[844],{"type":40,"tag":123,"props":845,"children":846},{"emptyLinePlaceholder":172},[847],{"type":45,"value":175},{"type":40,"tag":123,"props":849,"children":850},{"class":125,"line":312},[851],{"type":40,"tag":123,"props":852,"children":853},{},[854],{"type":45,"value":855},"documents\u002F\u003Ccorpus>\u002F\u003Cdoc-slug>\u002Fchunks\u002Fchunk-0001.md\n",{"type":40,"tag":123,"props":857,"children":858},{"class":125,"line":329},[859],{"type":40,"tag":123,"props":860,"children":861},{},[862],{"type":45,"value":863},"  Chunk content with frontmatter metadata.\n",{"type":40,"tag":123,"props":865,"children":866},{"class":125,"line":342},[867],{"type":40,"tag":123,"props":868,"children":869},{"emptyLinePlaceholder":172},[870],{"type":45,"value":175},{"type":40,"tag":123,"props":872,"children":873},{"class":125,"line":350},[874],{"type":40,"tag":123,"props":875,"children":876},{},[877],{"type":45,"value":878},"documents\u002F\u003Ccorpus>\u002Fchunks.jsonl\n",{"type":40,"tag":123,"props":880,"children":881},{"class":125,"line":372},[882],{"type":40,"tag":123,"props":883,"children":884},{},[885],{"type":45,"value":886},"  Machine-readable chunk export for custom indexing or debugging.\n",{"type":40,"tag":105,"props":888,"children":890},{"id":889},"old-api-mapping",[891],{"type":45,"value":892},"Old API mapping",{"type":40,"tag":894,"props":895,"children":896},"table",{},[897,916],{"type":40,"tag":898,"props":899,"children":900},"thead",{},[901],{"type":40,"tag":902,"props":903,"children":904},"tr",{},[905,911],{"type":40,"tag":906,"props":907,"children":908},"th",{},[909],{"type":45,"value":910},"Old Filesystem",{"type":40,"tag":906,"props":912,"children":913},{},[914],{"type":45,"value":915},"MemFS mimic",{"type":40,"tag":917,"props":918,"children":919},"tbody",{},[920,950,971,998,1015,1032,1061],{"type":40,"tag":902,"props":921,"children":922},{},[923,933],{"type":40,"tag":924,"props":925,"children":926},"td",{},[927],{"type":40,"tag":74,"props":928,"children":930},{"className":929},[],[931],{"type":45,"value":932},"folders.create",{"type":40,"tag":924,"props":934,"children":935},{},[936,942,944],{"type":40,"tag":74,"props":937,"children":939},{"className":938},[],[940],{"type":45,"value":941},"--corpus \u003Cname>",{"type":45,"value":943}," creates ",{"type":40,"tag":74,"props":945,"children":947},{"className":946},[],[948],{"type":45,"value":949},"documents\u002F\u003Ccorpus>\u002F",{"type":40,"tag":902,"props":951,"children":952},{},[953,962],{"type":40,"tag":924,"props":954,"children":955},{},[956],{"type":40,"tag":74,"props":957,"children":959},{"className":958},[],[960],{"type":45,"value":961},"folders.files.upload",{"type":40,"tag":924,"props":963,"children":964},{},[965],{"type":40,"tag":74,"props":966,"children":968},{"className":967},[],[969],{"type":45,"value":970},"letta_fs_to_memfs.py ingest --source \u003Cfile-or-directory-or-url>",{"type":40,"tag":902,"props":972,"children":973},{},[974,979],{"type":40,"tag":924,"props":975,"children":976},{},[977],{"type":45,"value":978},"OCR\u002Fchunk\u002Fembed job",{"type":40,"tag":924,"props":980,"children":981},{},[982,984,990,991,996],{"type":45,"value":983},"Extract + chunk locally; ",{"type":40,"tag":74,"props":985,"children":987},{"className":986},[],[988],{"type":45,"value":989},"qmd setup",{"type":45,"value":750},{"type":40,"tag":74,"props":992,"children":994},{"className":993},[],[995],{"type":45,"value":740},{"type":45,"value":997}," for semantic search",{"type":40,"tag":902,"props":999,"children":1000},{},[1001,1010],{"type":40,"tag":924,"props":1002,"children":1003},{},[1004],{"type":40,"tag":74,"props":1005,"children":1007},{"className":1006},[],[1008],{"type":45,"value":1009},"agents.folders.attach",{"type":40,"tag":924,"props":1011,"children":1012},{},[1013],{"type":45,"value":1014},"Enable MemFS, then review and sync repo changes when appropriate",{"type":40,"tag":902,"props":1016,"children":1017},{},[1018,1027],{"type":40,"tag":924,"props":1019,"children":1020},{},[1021],{"type":40,"tag":74,"props":1022,"children":1024},{"className":1023},[],[1025],{"type":45,"value":1026},"open_file",{"type":40,"tag":924,"props":1028,"children":1029},{},[1030],{"type":45,"value":1031},"Read chunk markdown files by path",{"type":40,"tag":902,"props":1033,"children":1034},{},[1035,1044],{"type":40,"tag":924,"props":1036,"children":1037},{},[1038],{"type":40,"tag":74,"props":1039,"children":1041},{"className":1040},[],[1042],{"type":45,"value":1043},"grep_file",{"type":40,"tag":924,"props":1045,"children":1046},{},[1047,1053,1055],{"type":40,"tag":74,"props":1048,"children":1050},{"className":1049},[],[1051],{"type":45,"value":1052},"rg",{"type":45,"value":1054}," or ",{"type":40,"tag":74,"props":1056,"children":1058},{"className":1057},[],[1059],{"type":45,"value":1060},"letta_fs_to_memfs.py search",{"type":40,"tag":902,"props":1062,"children":1063},{},[1064,1073],{"type":40,"tag":924,"props":1065,"children":1066},{},[1067],{"type":40,"tag":74,"props":1068,"children":1070},{"className":1069},[],[1071],{"type":45,"value":1072},"search_file",{"type":40,"tag":924,"props":1074,"children":1075},{},[1076,1082],{"type":40,"tag":74,"props":1077,"children":1079},{"className":1078},[],[1080],{"type":45,"value":1081},"letta_fs_to_memfs.py qmd query",{"type":45,"value":1083}," over the corpus chunk collection",{"type":40,"tag":105,"props":1085,"children":1087},{"id":1086},"heuristics",[1088],{"type":45,"value":1089},"Heuristics",{"type":40,"tag":1091,"props":1092,"children":1093},"ul",{},[1094,1105,1118,1137,1201,1249,1270,1275,1280,1293],{"type":40,"tag":63,"props":1095,"children":1096},{},[1097,1098,1103],{"type":45,"value":734},{"type":40,"tag":74,"props":1099,"children":1101},{"className":1100},[],[1102],{"type":45,"value":92},{"type":45,"value":1104}," for the small always-visible index only.",{"type":40,"tag":63,"props":1106,"children":1107},{},[1108,1110,1116],{"type":45,"value":1109},"Do not pin full documents into ",{"type":40,"tag":74,"props":1111,"children":1113},{"className":1112},[],[1114],{"type":45,"value":1115},"system\u002F",{"type":45,"value":1117},"; it will bloat the prompt.",{"type":40,"tag":63,"props":1119,"children":1120},{},[1121,1123,1128,1130,1136],{"type":45,"value":1122},"Keep chunks outside ",{"type":40,"tag":74,"props":1124,"children":1126},{"className":1125},[],[1127],{"type":45,"value":1115},{"type":45,"value":1129},", usually under ",{"type":40,"tag":74,"props":1131,"children":1133},{"className":1132},[],[1134],{"type":45,"value":1135},"documents\u002F\u003Ccorpus>\u002F...",{"type":45,"value":81},{"type":40,"tag":63,"props":1138,"children":1139},{},[1140,1142,1148,1150,1156,1158,1164,1165,1171,1172,1178,1179,1185,1186,1192,1193,1199],{"type":45,"value":1141},"Passing a directory to ",{"type":40,"tag":74,"props":1143,"children":1145},{"className":1144},[],[1146],{"type":45,"value":1147},"--source",{"type":45,"value":1149}," recursively ingests supported files (",{"type":40,"tag":74,"props":1151,"children":1153},{"className":1152},[],[1154],{"type":45,"value":1155},".pdf",{"type":45,"value":1157},", ",{"type":40,"tag":74,"props":1159,"children":1161},{"className":1160},[],[1162],{"type":45,"value":1163},".md",{"type":45,"value":1157},{"type":40,"tag":74,"props":1166,"children":1168},{"className":1167},[],[1169],{"type":45,"value":1170},".txt",{"type":45,"value":1157},{"type":40,"tag":74,"props":1173,"children":1175},{"className":1174},[],[1176],{"type":45,"value":1177},".json",{"type":45,"value":1157},{"type":40,"tag":74,"props":1180,"children":1182},{"className":1181},[],[1183],{"type":45,"value":1184},".csv",{"type":45,"value":1157},{"type":40,"tag":74,"props":1187,"children":1189},{"className":1188},[],[1190],{"type":45,"value":1191},".html",{"type":45,"value":1157},{"type":40,"tag":74,"props":1194,"children":1196},{"className":1195},[],[1197],{"type":45,"value":1198},".xml",{"type":45,"value":1200},").",{"type":40,"tag":63,"props":1202,"children":1203},{},[1204,1205,1211,1212,1218,1220,1226,1227,1233,1234,1240,1242,1248],{"type":45,"value":734},{"type":40,"tag":74,"props":1206,"children":1208},{"className":1207},[],[1209],{"type":45,"value":1210},"--glob",{"type":45,"value":750},{"type":40,"tag":74,"props":1213,"children":1215},{"className":1214},[],[1216],{"type":45,"value":1217},"--exclude",{"type":45,"value":1219}," for messy directories. Defaults exclude ",{"type":40,"tag":74,"props":1221,"children":1223},{"className":1222},[],[1224],{"type":45,"value":1225},".git",{"type":45,"value":1157},{"type":40,"tag":74,"props":1228,"children":1230},{"className":1229},[],[1231],{"type":45,"value":1232},"node_modules",{"type":45,"value":1157},{"type":40,"tag":74,"props":1235,"children":1237},{"className":1236},[],[1238],{"type":45,"value":1239},".venv",{"type":45,"value":1241},", and ",{"type":40,"tag":74,"props":1243,"children":1245},{"className":1244},[],[1246],{"type":45,"value":1247},"__pycache__",{"type":45,"value":81},{"type":40,"tag":63,"props":1250,"children":1251},{},[1252,1254,1260,1262,1268],{"type":45,"value":1253},"URL downloads default to ",{"type":40,"tag":74,"props":1255,"children":1257},{"className":1256},[],[1258],{"type":45,"value":1259},"--max-download-mb 100",{"type":45,"value":1261},"; set ",{"type":40,"tag":74,"props":1263,"children":1265},{"className":1264},[],[1266],{"type":45,"value":1267},"0",{"type":45,"value":1269}," for unlimited.",{"type":40,"tag":63,"props":1271,"children":1272},{},[1273],{"type":45,"value":1274},"Re-ingesting the same document slug replaces that document's old chunk directory, so stale chunks do not survive chunk-size changes.",{"type":40,"tag":63,"props":1276,"children":1277},{},[1278],{"type":45,"value":1279},"Use chunk sizes around 2,000-4,000 characters with 200-500 character overlap.",{"type":40,"tag":63,"props":1281,"children":1282},{},[1283,1285,1291],{"type":45,"value":1284},"Use the CLI's ",{"type":40,"tag":74,"props":1286,"children":1288},{"className":1287},[],[1289],{"type":45,"value":1290},"qmd",{"type":45,"value":1292}," subcommands when the user needs semantic search over many chunks.",{"type":40,"tag":63,"props":1294,"children":1295},{},[1296],{"type":45,"value":1297},"Preserve source URLs, checksums, page markers, chunk numbers, and corpus names in the generated files.",{"type":40,"tag":105,"props":1299,"children":1301},{"id":1300},"cli-reference",[1302],{"type":45,"value":1303},"CLI reference",{"type":40,"tag":112,"props":1305,"children":1307},{"className":114,"code":1306,"language":116,"meta":117,"style":117},"uv run --with pymupdf \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" ingest --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" search --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd setup --help\nuv run \"$SKILL_DIR\u002Fscripts\u002Fletta_fs_to_memfs.py\" qmd query --help\n",[1308],{"type":40,"tag":74,"props":1309,"children":1310},{"__ignoreMap":117},[1311,1355,1390,1429],{"type":40,"tag":123,"props":1312,"children":1313},{"class":125,"line":126},[1314,1318,1322,1326,1330,1334,1338,1342,1346,1350],{"type":40,"tag":123,"props":1315,"children":1316},{"style":191},[1317],{"type":45,"value":194},{"type":40,"tag":123,"props":1319,"children":1320},{"style":157},[1321],{"type":45,"value":199},{"type":40,"tag":123,"props":1323,"children":1324},{"style":157},[1325],{"type":45,"value":204},{"type":40,"tag":123,"props":1327,"children":1328},{"style":157},[1329],{"type":45,"value":209},{"type":40,"tag":123,"props":1331,"children":1332},{"style":146},[1333],{"type":45,"value":260},{"type":40,"tag":123,"props":1335,"children":1336},{"style":140},[1337],{"type":45,"value":228},{"type":40,"tag":123,"props":1339,"children":1340},{"style":157},[1341],{"type":45,"value":233},{"type":40,"tag":123,"props":1343,"children":1344},{"style":146},[1345],{"type":45,"value":154},{"type":40,"tag":123,"props":1347,"children":1348},{"style":157},[1349],{"type":45,"value":242},{"type":40,"tag":123,"props":1351,"children":1352},{"style":157},[1353],{"type":45,"value":1354}," --help\n",{"type":40,"tag":123,"props":1356,"children":1357},{"class":125,"line":136},[1358,1362,1366,1370,1374,1378,1382,1386],{"type":40,"tag":123,"props":1359,"children":1360},{"style":191},[1361],{"type":45,"value":194},{"type":40,"tag":123,"props":1363,"children":1364},{"style":157},[1365],{"type":45,"value":199},{"type":40,"tag":123,"props":1367,"children":1368},{"style":146},[1369],{"type":45,"value":260},{"type":40,"tag":123,"props":1371,"children":1372},{"style":140},[1373],{"type":45,"value":228},{"type":40,"tag":123,"props":1375,"children":1376},{"style":157},[1377],{"type":45,"value":233},{"type":40,"tag":123,"props":1379,"children":1380},{"style":146},[1381],{"type":45,"value":154},{"type":40,"tag":123,"props":1383,"children":1384},{"style":157},[1385],{"type":45,"value":450},{"type":40,"tag":123,"props":1387,"children":1388},{"style":157},[1389],{"type":45,"value":1354},{"type":40,"tag":123,"props":1391,"children":1392},{"class":125,"line":168},[1393,1397,1401,1405,1409,1413,1417,1421,1425],{"type":40,"tag":123,"props":1394,"children":1395},{"style":191},[1396],{"type":45,"value":194},{"type":40,"tag":123,"props":1398,"children":1399},{"style":157},[1400],{"type":45,"value":199},{"type":40,"tag":123,"props":1402,"children":1403},{"style":146},[1404],{"type":45,"value":260},{"type":40,"tag":123,"props":1406,"children":1407},{"style":140},[1408],{"type":45,"value":228},{"type":40,"tag":123,"props":1410,"children":1411},{"style":157},[1412],{"type":45,"value":233},{"type":40,"tag":123,"props":1414,"children":1415},{"style":146},[1416],{"type":45,"value":154},{"type":40,"tag":123,"props":1418,"children":1419},{"style":157},[1420],{"type":45,"value":570},{"type":40,"tag":123,"props":1422,"children":1423},{"style":157},[1424],{"type":45,"value":575},{"type":40,"tag":123,"props":1426,"children":1427},{"style":157},[1428],{"type":45,"value":1354},{"type":40,"tag":123,"props":1430,"children":1431},{"class":125,"line":178},[1432,1436,1440,1444,1448,1452,1456,1460,1464],{"type":40,"tag":123,"props":1433,"children":1434},{"style":191},[1435],{"type":45,"value":194},{"type":40,"tag":123,"props":1437,"children":1438},{"style":157},[1439],{"type":45,"value":199},{"type":40,"tag":123,"props":1441,"children":1442},{"style":146},[1443],{"type":45,"value":260},{"type":40,"tag":123,"props":1445,"children":1446},{"style":140},[1447],{"type":45,"value":228},{"type":40,"tag":123,"props":1449,"children":1450},{"style":157},[1451],{"type":45,"value":233},{"type":40,"tag":123,"props":1453,"children":1454},{"style":146},[1455],{"type":45,"value":154},{"type":40,"tag":123,"props":1457,"children":1458},{"style":157},[1459],{"type":45,"value":570},{"type":40,"tag":123,"props":1461,"children":1462},{"style":157},[1463],{"type":45,"value":657},{"type":40,"tag":123,"props":1465,"children":1466},{"style":157},[1467],{"type":45,"value":1354},{"type":40,"tag":48,"props":1469,"children":1470},{},[1471],{"type":45,"value":1472},"Compatibility wrappers remain for older examples:",{"type":40,"tag":112,"props":1474,"children":1476},{"className":114,"code":1475,"language":116,"meta":117,"style":117},"uv run --with pymupdf \"$SKILL_DIR\u002Fscripts\u002Fingest_documents.py\" --memory-dir \"$MEMORY_DIR\" --corpus docs --source .\u002Fdocs\nuv run \"$SKILL_DIR\u002Fscripts\u002Fsearch_corpus.py\" --memory-dir \"$MEMORY_DIR\" --corpus docs --query \"refund policy\"\n",[1477],{"type":40,"tag":74,"props":1478,"children":1479},{"__ignoreMap":117},[1480,1553],{"type":40,"tag":123,"props":1481,"children":1482},{"class":125,"line":126},[1483,1487,1491,1495,1499,1503,1507,1512,1516,1521,1525,1529,1533,1538,1543,1548],{"type":40,"tag":123,"props":1484,"children":1485},{"style":191},[1486],{"type":45,"value":194},{"type":40,"tag":123,"props":1488,"children":1489},{"style":157},[1490],{"type":45,"value":199},{"type":40,"tag":123,"props":1492,"children":1493},{"style":157},[1494],{"type":45,"value":204},{"type":40,"tag":123,"props":1496,"children":1497},{"style":157},[1498],{"type":45,"value":209},{"type":40,"tag":123,"props":1500,"children":1501},{"style":146},[1502],{"type":45,"value":260},{"type":40,"tag":123,"props":1504,"children":1505},{"style":140},[1506],{"type":45,"value":228},{"type":40,"tag":123,"props":1508,"children":1509},{"style":157},[1510],{"type":45,"value":1511},"\u002Fscripts\u002Fingest_documents.py",{"type":40,"tag":123,"props":1513,"children":1514},{"style":146},[1515],{"type":45,"value":154},{"type":40,"tag":123,"props":1517,"children":1518},{"style":157},[1519],{"type":45,"value":1520}," --memory-dir",{"type":40,"tag":123,"props":1522,"children":1523},{"style":146},[1524],{"type":45,"value":260},{"type":40,"tag":123,"props":1526,"children":1527},{"style":140},[1528],{"type":45,"value":265},{"type":40,"tag":123,"props":1530,"children":1531},{"style":146},[1532],{"type":45,"value":154},{"type":40,"tag":123,"props":1534,"children":1535},{"style":157},[1536],{"type":45,"value":1537}," --corpus",{"type":40,"tag":123,"props":1539,"children":1540},{"style":157},[1541],{"type":45,"value":1542}," docs",{"type":40,"tag":123,"props":1544,"children":1545},{"style":157},[1546],{"type":45,"value":1547}," --source",{"type":40,"tag":123,"props":1549,"children":1550},{"style":157},[1551],{"type":45,"value":1552}," .\u002Fdocs\n",{"type":40,"tag":123,"props":1554,"children":1555},{"class":125,"line":136},[1556,1560,1564,1568,1572,1577,1581,1585,1589,1593,1597,1601,1605,1610,1614,1619],{"type":40,"tag":123,"props":1557,"children":1558},{"style":191},[1559],{"type":45,"value":194},{"type":40,"tag":123,"props":1561,"children":1562},{"style":157},[1563],{"type":45,"value":199},{"type":40,"tag":123,"props":1565,"children":1566},{"style":146},[1567],{"type":45,"value":260},{"type":40,"tag":123,"props":1569,"children":1570},{"style":140},[1571],{"type":45,"value":228},{"type":40,"tag":123,"props":1573,"children":1574},{"style":157},[1575],{"type":45,"value":1576},"\u002Fscripts\u002Fsearch_corpus.py",{"type":40,"tag":123,"props":1578,"children":1579},{"style":146},[1580],{"type":45,"value":154},{"type":40,"tag":123,"props":1582,"children":1583},{"style":157},[1584],{"type":45,"value":1520},{"type":40,"tag":123,"props":1586,"children":1587},{"style":146},[1588],{"type":45,"value":260},{"type":40,"tag":123,"props":1590,"children":1591},{"style":140},[1592],{"type":45,"value":265},{"type":40,"tag":123,"props":1594,"children":1595},{"style":146},[1596],{"type":45,"value":154},{"type":40,"tag":123,"props":1598,"children":1599},{"style":157},[1600],{"type":45,"value":1537},{"type":40,"tag":123,"props":1602,"children":1603},{"style":157},[1604],{"type":45,"value":1542},{"type":40,"tag":123,"props":1606,"children":1607},{"style":157},[1608],{"type":45,"value":1609}," --query",{"type":40,"tag":123,"props":1611,"children":1612},{"style":146},[1613],{"type":45,"value":260},{"type":40,"tag":123,"props":1615,"children":1616},{"style":157},[1617],{"type":45,"value":1618},"refund policy",{"type":40,"tag":123,"props":1620,"children":1621},{"style":146},[1622],{"type":45,"value":165},{"type":40,"tag":105,"props":1624,"children":1626},{"id":1625},"pdf-notes",[1627],{"type":45,"value":1628},"PDF notes",{"type":40,"tag":48,"props":1630,"children":1631},{},[1632,1634,1640],{"type":45,"value":1633},"The ingest script uses PyMuPDF when it sees a PDF. Run it with ",{"type":40,"tag":74,"props":1635,"children":1637},{"className":1636},[],[1638],{"type":45,"value":1639},"uv run --with pymupdf ...",{"type":45,"value":81},{"type":40,"tag":48,"props":1642,"children":1643},{},[1644,1646,1652],{"type":45,"value":1645},"For scanned PDFs or complex tables, use the ",{"type":40,"tag":74,"props":1647,"children":1649},{"className":1648},[],[1650],{"type":45,"value":1651},"tools\u002Fextracting-pdf-text",{"type":45,"value":1653}," skill first, then ingest the extracted markdown\u002Ftext file with this skill.",{"type":40,"tag":48,"props":1655,"children":1656},{},[1657,1659,1665],{"type":45,"value":1658},"See ",{"type":40,"tag":74,"props":1660,"children":1662},{"className":1661},[],[1663],{"type":45,"value":1664},"references\u002Fdesign.md",{"type":45,"value":1666}," for design notes and edge cases.",{"type":40,"tag":1668,"props":1669,"children":1670},"style",{},[1671],{"type":45,"value":1672},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1674,"total":1835},[1675,1693,1708,1720,1732,1746,1758,1769,1781,1797,1808,1820],{"slug":1676,"name":1676,"fn":1677,"description":1678,"org":1679,"tags":1680,"stars":1690,"repoUrl":1691,"updatedAt":1692},"acquiring-skills","discover and install agent skills","Discover and install skills from Hermes, ClawHub, GitHub, and other registries. Load this skill whenever a user asks for a capability you don't already have — image generation, social media, email, calendar, finance, DevOps, search, browser automation, etc.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1681,1684,1687],{"name":1682,"slug":1683,"type":16},"Agents","agents",{"name":1685,"slug":1686,"type":16},"Automation","automation",{"name":1688,"slug":1689,"type":16},"GitHub","github",2831,"https:\u002F\u002Fgithub.com\u002Fletta-ai\u002Fletta-code","2026-07-13T06:22:58.45767",{"slug":1694,"name":1695,"fn":1696,"description":1697,"org":1698,"tags":1699,"stars":1690,"repoUrl":1691,"updatedAt":1707},"context-doctor","Context Doctor","repair system prompt and memory degradation","Identify and repair degradation in system prompt, external memory, and skills preventing you from following instructions or remembering information as well as you should.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1700,1701,1704],{"name":1682,"slug":1683,"type":16},{"name":1702,"slug":1703,"type":16},"AI Context","ai-context",{"name":1705,"slug":1706,"type":16},"Debugging","debugging","2026-07-13T06:22:50.151002",{"slug":1709,"name":1709,"fn":1710,"description":1711,"org":1712,"tags":1713,"stars":1690,"repoUrl":1691,"updatedAt":1719},"converting-mcps-to-skills","connect MCP servers to create skills","Connect to MCP (Model Context Protocol) servers and create skills for repeated use. Load when a user wants to use an MCP server, connect to external tools via MCP, or when they mention MCP, model context protocol, or specific MCP servers.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1714,1715,1716],{"name":1682,"slug":1683,"type":16},{"name":1685,"slug":1686,"type":16},{"name":1717,"slug":1718,"type":16},"MCP","mcp","2026-07-13T06:23:37.646079",{"slug":1721,"name":1721,"fn":1722,"description":1723,"org":1724,"tags":1725,"stars":1690,"repoUrl":1691,"updatedAt":1731},"creating-mods","create and edit Letta Code mods","Creates and edits trusted local Letta Code mods, including tools, slash commands, local-only model providers, lifecycle\u002Fturn events, scoped conversation helpers, panels, and capability-gated behavior. Use when asked to make a mod, add an agent-callable tool, add a slash command, add a local provider\u002Fmodel adapter, transform turns, react to app events, or add lightweight mod UI outside the dedicated \u002Fstatusline flow.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1726,1727,1728],{"name":1682,"slug":1683,"type":16},{"name":1685,"slug":1686,"type":16},{"name":1729,"slug":1730,"type":16},"Coding","coding","2026-07-23T05:42:38.133565",{"slug":1733,"name":1733,"fn":1734,"description":1735,"org":1736,"tags":1737,"stars":1690,"repoUrl":1691,"updatedAt":1745},"creating-skills","create and update agent skills","Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Letta Code's capabilities with specialized knowledge, workflows, or tool integrations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1738,1739,1742],{"name":1682,"slug":1683,"type":16},{"name":1740,"slug":1741,"type":16},"Documentation","documentation",{"name":1743,"slug":1744,"type":16},"Plugin Development","plugin-development","2026-07-13T06:22:56.998659",{"slug":1747,"name":1747,"fn":1748,"description":1749,"org":1750,"tags":1751,"stars":1690,"repoUrl":1691,"updatedAt":1757},"customizing-commands","create and manage Letta slash commands","Creates, edits, and enables Letta Code mod-provided slash commands. Use when the user asks to add a custom \u002Fcommand, slash command, command shortcut, scoped conversation-backed command, or command-driven panel behavior.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1752,1753,1754],{"name":1682,"slug":1683,"type":16},{"name":1685,"slug":1686,"type":16},{"name":1755,"slug":1756,"type":16},"CLI","cli","2026-07-13T06:23:18.266798",{"slug":1759,"name":1759,"fn":1760,"description":1761,"org":1762,"tags":1763,"stars":1690,"repoUrl":1691,"updatedAt":1768},"customizing-statusline","customize Letta Code statusline mods","Creates, edits, and migrates Letta Code statusline mods. Use when handling the \u002Fstatusline command or continuing work started by \u002Fstatusline.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1764,1765],{"name":1755,"slug":1756,"type":16},{"name":1766,"slug":1767,"type":16},"Engineering","engineering","2026-07-13T06:23:27.465985",{"slug":1770,"name":1770,"fn":1771,"description":1772,"org":1773,"tags":1774,"stars":1690,"repoUrl":1691,"updatedAt":1780},"dispatching-coding-agents","dispatch stateless coding agents","Dispatch stateless coding agents (Claude Code or Codex) via Bash. Use when you're stuck, need a second opinion, or need parallel research on a hard problem. They have no memory — you must provide all context.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1775,1776,1777],{"name":1682,"slug":1683,"type":16},{"name":1729,"slug":1730,"type":16},{"name":1778,"slug":1779,"type":16},"Multi-Agent","multi-agent","2026-07-26T05:46:56.388845",{"slug":1782,"name":1782,"fn":1783,"description":1784,"org":1785,"tags":1786,"stars":1690,"repoUrl":1691,"updatedAt":1796},"editing-letta-code-desktop-preferences","edit Letta Code Desktop preferences","Edits Letta Code Desktop (LCD) preferences by safely reading and updating ~\u002F.letta\u002Fdesktop_preferences.json. Use only when the user asks to change current Desktop\u002FLCD settings such as theme, default working directory, remote access preference, or remote environment name via the preferences JSON.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1787,1790,1793],{"name":1788,"slug":1789,"type":16},"Configuration","configuration",{"name":1791,"slug":1792,"type":16},"Desktop","desktop",{"name":1794,"slug":1795,"type":16},"Themes","themes","2026-07-13T06:23:41.407811",{"slug":1798,"name":1798,"fn":1799,"description":1800,"org":1801,"tags":1802,"stars":1690,"repoUrl":1691,"updatedAt":1807},"finding-agents","locate and manage agents on server","Find other agents on the same server. Use when the user asks about other agents, wants to migrate memory from another agent, or needs to find an agent by name or tags.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1803,1804],{"name":1682,"slug":1683,"type":16},{"name":1805,"slug":1806,"type":16},"Management","management","2026-07-16T06:02:17.297841",{"slug":1809,"name":1809,"fn":1810,"description":1811,"org":1812,"tags":1813,"stars":1690,"repoUrl":1691,"updatedAt":1819},"generating-mod-envs","generate Letta mod learning environments","Generates and reviews mod learning env JSON files for Letta Code local mods. Use when asked to teach, learn, or optimize a mod behavior; create, draft, validate, improve, or explain envs for `\u002Fmods learn --env`; or design evaluation scenarios, memory fixtures, requiredResultMarkers, requiredTraceMarkers, negative controls, and candidate diversity hints.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1814,1815,1818],{"name":1682,"slug":1683,"type":16},{"name":1816,"slug":1817,"type":16},"AI Infrastructure","ai-infrastructure",{"name":1788,"slug":1789,"type":16},"2026-07-13T06:23:08.838181",{"slug":1821,"name":1821,"fn":1822,"description":1823,"org":1824,"tags":1825,"stars":1690,"repoUrl":1691,"updatedAt":1834},"image-generation","generate images from text prompts","Generate images from text prompts (and optionally edit\u002Fremix input images). Use when the user asks to create, generate, draw, render, or edit an image, illustration, logo, icon, diagram, or photo.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1826,1829,1832],{"name":1827,"slug":1828,"type":16},"Creative","creative",{"name":1830,"slug":1831,"type":16},"Graphics","graphics",{"name":1833,"slug":1821,"type":16},"Image Generation","2026-07-13T06:23:06.189403",69,{"items":1837,"total":1949},[1838,1852,1867,1886,1899,1920,1930],{"slug":1839,"name":1839,"fn":1840,"description":1841,"org":1842,"tags":1843,"stars":23,"repoUrl":24,"updatedAt":1851},"1password","manage secrets with 1Password CLI","Set up and use 1Password CLI (op). Use when installing the CLI, enabling desktop app integration, signing in (single or multi-account), or reading\u002Finjecting\u002Frunning secrets via op.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1844,1847,1848],{"name":1845,"slug":1846,"type":16},"Authentication","authentication",{"name":1755,"slug":1756,"type":16},{"name":1849,"slug":1850,"type":16},"Security","security","2026-07-13T06:24:39.504387",{"slug":1853,"name":1853,"fn":1854,"description":1855,"org":1856,"tags":1857,"stars":23,"repoUrl":24,"updatedAt":1866},"agent-slack","automate Slack messaging and workflows","Slack automation CLI — read\u002Fsend\u002Fsearch messages, browse threads and channels, manage channels, download attachments, look up users, and run workflows.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1858,1859,1860,1863],{"name":1685,"slug":1686,"type":16},{"name":1755,"slug":1756,"type":16},{"name":1861,"slug":1862,"type":16},"Messaging","messaging",{"name":1864,"slug":1865,"type":16},"Slack","slack","2026-07-13T06:23:51.908511",{"slug":1868,"name":1868,"fn":1869,"description":1870,"org":1871,"tags":1872,"stars":23,"repoUrl":24,"updatedAt":1885},"ai-news","fetch and summarize AI news","Fetch and summarize recent AI news from curated RSS feeds (Hugging Face, VentureBeat, The Verge, OpenAI, Anthropic, DeepMind, etc.) and YouTube channels (Yannic Kilcher, Two Minute Papers, AI Explained, etc.). Also fetches full transcripts for specific YouTube videos. Use when the user asks about recent AI news, what's happened in AI lately, summaries of AI research or product announcements, or wants a digest of what's going on in the AI space.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1873,1876,1879,1882],{"name":1874,"slug":1875,"type":16},"Communications","communications",{"name":1877,"slug":1878,"type":16},"LLM","llm",{"name":1880,"slug":1881,"type":16},"Research","research",{"name":1883,"slug":1884,"type":16},"Summarization","summarization","2026-07-13T06:24:20.520223",{"slug":1887,"name":1887,"fn":1888,"description":1889,"org":1890,"tags":1891,"stars":23,"repoUrl":24,"updatedAt":1898},"creating-letta-code-channels","build and debug Letta Code channels","Builds and debugs Letta Code channels, including first-party channel adapters and dynamic user channel plugins under ~\u002F.letta\u002Fchannels. Use when adding Telegram, WhatsApp, Bluesky, Slack, Discord, or custom channel support; testing channel routing, pairing, MessageChannel, runtime dependencies, or channel plugin manifests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1892,1893,1896,1897],{"name":1682,"slug":1683,"type":16},{"name":1894,"slug":1895,"type":16},"API Development","api-development",{"name":1861,"slug":1862,"type":16},{"name":1864,"slug":1865,"type":16},"2026-07-13T06:25:55.843495",{"slug":1900,"name":1900,"fn":1901,"description":1902,"org":1903,"tags":1904,"stars":23,"repoUrl":24,"updatedAt":1919},"datadog","query Datadog observability data","Query Datadog observability data (logs, metrics, monitors, dashboards, hosts) via direct API. Use when investigating production issues, checking monitors, searching logs, or accessing Datadog data.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1905,1907,1910,1913,1916],{"name":1906,"slug":1900,"type":16},"Datadog",{"name":1908,"slug":1909,"type":16},"Logs","logs",{"name":1911,"slug":1912,"type":16},"Metrics","metrics",{"name":1914,"slug":1915,"type":16},"Monitoring","monitoring",{"name":1917,"slug":1918,"type":16},"Observability","observability","2026-07-13T06:24:27.990605",{"slug":1921,"name":1921,"fn":1922,"description":1923,"org":1924,"tags":1925,"stars":23,"repoUrl":24,"updatedAt":1929},"discord","automate Discord server and channel tasks","Discord automation CLI — send\u002Fread\u002Fsearch messages, manage channels and servers, react, create threads, pin messages, and look up users.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1926,1927,1928],{"name":1685,"slug":1686,"type":16},{"name":1755,"slug":1756,"type":16},{"name":1861,"slug":1862,"type":16},"2026-07-13T06:24:26.62387",{"slug":1931,"name":1931,"fn":1932,"description":1933,"org":1934,"tags":1935,"stars":23,"repoUrl":24,"updatedAt":1948},"doc","create and edit Word documents","Use when the task involves reading, creating, or editing `.docx` documents, especially when formatting or layout fidelity matters; prefer `python-docx` plus the bundled `scripts\u002Frender_docx.py` for visual checks.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1936,1939,1942,1945],{"name":1937,"slug":1938,"type":16},"Documents","documents",{"name":1940,"slug":1941,"type":16},"DOCX","docx",{"name":1943,"slug":1944,"type":16},"Office","office",{"name":1946,"slug":1947,"type":16},"Word","word","2026-07-13T06:23:44.299568",45]