{"id":"d483a8f6-cedf-42e7-9b7e-ffc3fe66b2b9","shortId":"gpazRv","kind":"skill","title":"agent-sheet","tagline":"Deprecated legacy alias for `univer-cli`; use only for older prompts or installs that still call `agent-sheet`. Prefer `univer-cli` for current shell-native workbook work.","description":"# agent-sheet\n\nDeprecated: `agent-sheet` is the old name for `univer-cli`. Prefer `univer-cli` and the `univer` / `unv` command surface for new work.\n\nKeep using this skill only when an older prompt, install, or environment explicitly refers to `agent-sheet`.\n\n`agent-sheet` is the shell-native workbook CLI for agent work. Use it when the task needs real workbook structure, formulas, workbook-local logic, or a safe stdin/stdout roundtrip instead of ad hoc CSV handling.\n\nTreat the workbook as the source of truth. Read from workbook-visible state, mutate through the public surface, and prove success from workbook-visible results.\n\n## Start here\n\nUse the smallest surface that cleanly matches the task:\n\n- `inspect`: first read, reconnaissance, workbook-visible structure, formulas, ranges, lint signals\n- `search`: first-class localization before edits\n- `fill`: first-class propagation when a correct seed already exists\n- `run`: default programmable workbook surface for bounded workbook-local logic\n- `pipe out` / `pipe in`: bulk rectangular data plane for shell roundtrips\n\nStart small. If `inspect`, `search`, `fill`, or `pipe` cleanly expresses the job, use that surface directly. Use `run` when the work is programmable workbook logic rather than bulk data movement.\n\n## Default operating model\n\n- if you do not have an `entry-id` yet, get one first with `file list`, `file create`, or `file import`, then stay on that workbook\n- resolve one workbook and keep its `--entry-id` for the whole task\n- start with `inspect` so worksheet names, headers, formulas, and write boundaries are real rather than guessed\n- use `search` before mutation when the target is content-defined\n- use `fill` for spreadsheet-native propagation from a known seed\n- use `pipe out` / `pipe in` when the shell is the right place for a rectangular transform\n- use `run` for bounded workbook-local logic, structural edits, formatting, or formula orchestration\n- verify from workbook-visible reads after every mutation or handoff step\n\n## Hard defaults\n\n- prefer `--entry-id` to keep the target explicit\n- start with `inspect`\n- treat `run` as the default programmable workbook surface\n- treat `pipe out` / `pipe in` as the bulk rectangular data plane\n- verify from workbook-visible results, not metadata alone\n- verify shell roundtrips with structure plus sample rows, not row count alone\n- use workbook-visible inspection for handoff and completion checks\n\n## Highest-signal gotchas\n\n- imported local entries can be healthy even when later `file info` shows `unitId: null`; keep operating on `entryId`\n- `file info` is metadata only; it does not prove worksheet count, worksheet names, or formula state\n- non-English worksheet names work, but quote the full A1 range string in the shell\n- shell pipelines can preserve the expected row count while still shifting headers or keys; verify the first rows and key columns after writeback\n- if workbook-visible verification disagrees with metadata, trust the workbook-visible surface first\n\n## Read next\n\nRead only what matches the task:\n\n- [playbooks/01-getting-started.md](playbooks/01-getting-started.md): first contact, target resolution, basic flow\n- [playbooks/02-common-workflows.md](playbooks/02-common-workflows.md): workflow recipes by task intent\n- [playbooks/03-handoff-and-verification.md](playbooks/03-handoff-and-verification.md): handoff, completion checks, workbook-visible verification\n- [references/command-selection-matrix.md](references/command-selection-matrix.md): shortest path from task intent to command choice\n- [references/run-api.md](references/run-api.md): `run` entry point and reference map for bounded programmable workbook logic\n- [references/shell-patterns.md](references/shell-patterns.md): small shell-native patterns for `pipe` workflows\n- [references/gotchas.md](references/gotchas.md): real-world failure modes and edge cases\n- [examples/handoff-verify.md](examples/handoff-verify.md): export/import/handoff proof\n- [examples/roundtrip-pipe-review-rectangle.md](examples/roundtrip-pipe-review-rectangle.md): staged shell roundtrip with explicit preview verification","tags":["agent","sheet","skills","dream-num","agent-skills","claude-code-skills","claude-skills","cli","openai-codex","openclaw-skills","spreadsheet"],"capabilities":["skill","source-dream-num","skill-agent-sheet","topic-agent-skills","topic-claude-code-skills","topic-claude-skills","topic-cli","topic-openai-codex","topic-openclaw-skills","topic-skills","topic-spreadsheet"],"categories":["skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/dream-num/skills/agent-sheet","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add dream-num/skills","source_repo":"https://github.com/dream-num/skills","install_from":"skills.sh"}},"qualityScore":"0.457","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 14 github stars · SKILL.md body (4,134 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-14T01:04:28.573Z","embedding":null,"createdAt":"2026-04-23T13:03:52.146Z","updatedAt":"2026-05-14T01:04:28.573Z","lastSeenAt":"2026-05-14T01:04:28.573Z","tsv":"'a1':473 'ad':115 'agent':2,22,36,40,79,82,92 'agent-sheet':1,21,35,39,78,81 'alia':6 'alon':402,414 'alreadi':185 'basic':531 'bound':193,338,568 'boundari':291 'bulk':202,236,390 'call':20 'case':591 'check':424,544 'choic':558 'class':172,179 'clean':153,217 'cli':10,27,49,53,90 'column':499 'command':58,557 'complet':423,543 'contact':528 'content':306 'content-defin':305 'correct':183 'count':413,457,486 'creat':259 'csv':117 'current':29 'data':204,237,392 'default':188,239,362,379 'defin':307 'deprec':4,38 'direct':224 'disagre':507 'edg':590 'edit':175,344 'english':465 'entri':249,275,365,431,562 'entry-id':248,274,364 'entryid':446 'environ':74 'even':435 'everi':356 'examples/handoff-verify.md':592,593 'examples/roundtrip-pipe-review-rectangle.md':596,597 'exist':186 'expect':484 'explicit':75,371,602 'export/import/handoff':594 'express':218 'failur':587 'file':256,258,261,438,447 'fill':176,214,309 'first':158,171,178,254,495,516,527 'first-class':170,177 'flow':532 'format':345 'formula':103,165,288,347,461 'full':472 'get':252 'gotcha':428 'guess':296 'handl':118 'handoff':359,421,542 'hard':361 'header':287,490 'healthi':434 'highest':426 'highest-sign':425 'hoc':116 'id':250,276,366 'import':262,429 'info':439,448 'inspect':157,212,283,374,419 'instal':17,72 'instead':113 'intent':539,555 'job':220 'keep':63,272,368,443 'key':492,498 'known':317 'later':437 'legaci':5 'lint':167 'list':257 'local':106,173,196,341,430 'logic':107,197,233,342,571 'map':566 'match':154,522 'metadata':401,450,509 'mode':588 'model':241 'movement':238 'mutat':133,300,357 'name':45,286,459,467 'nativ':32,88,313,577 'need':99 'new':61 'next':518 'non':464 'non-english':463 'null':442 'old':44 'older':14,70 'one':253,269 'oper':240,444 'orchestr':348 'path':552 'pattern':578 'pipe':198,200,216,320,322,384,386,580 'pipelin':480 'place':330 'plane':205,393 'playbooks/01-getting-started.md':525,526 'playbooks/02-common-workflows.md':533,534 'playbooks/03-handoff-and-verification.md':540,541 'plus':408 'point':563 'prefer':24,50,363 'preserv':482 'preview':603 'programm':189,231,380,569 'prompt':15,71 'proof':595 'propag':180,314 'prove':139,455 'public':136 'quot':470 'rang':166,474 'rather':234,294 'read':127,159,354,517,519 'real':100,293,585 'real-world':584 'recip':536 'reconnaiss':160 'rectangular':203,333,391 'refer':76,565 'references/command-selection-matrix.md':549,550 'references/gotchas.md':582,583 'references/run-api.md':559,560 'references/shell-patterns.md':572,573 'resolut':530 'resolv':268 'result':145,399 'right':329 'roundtrip':112,208,405,600 'row':410,412,485,496 'run':187,226,336,376,561 'safe':110 'sampl':409 'search':169,213,298 'seed':184,318 'sheet':3,23,37,41,80,83 'shell':31,87,207,326,404,478,479,576,599 'shell-nat':30,86,575 'shift':489 'shortest':551 'show':440 'signal':168,427 'skill':66 'skill-agent-sheet' 'small':210,574 'smallest':150 'sourc':124 'source-dream-num' 'spreadsheet':312 'spreadsheet-n':311 'stage':598 'start':146,209,281,372 'state':132,462 'stay':264 'stdin/stdout':111 'step':360 'still':19,488 'string':475 'structur':102,164,343,407 'success':140 'surfac':59,137,151,191,223,382,515 'target':303,370,529 'task':98,156,280,524,538,554 'topic-agent-skills' 'topic-claude-code-skills' 'topic-claude-skills' 'topic-cli' 'topic-openai-codex' 'topic-openclaw-skills' 'topic-skills' 'topic-spreadsheet' 'transform':334 'treat':119,375,383 'trust':510 'truth':126 'unitid':441 'univ':9,26,48,52,56 'univer-c':8,25,47,51 'unv':57 'use':11,64,94,148,221,225,297,308,319,335,415 'verif':506,548,604 'verifi':349,394,403,493 'visibl':131,144,163,353,398,418,505,514,547 'whole':279 'work':34,62,93,229,468 'workbook':33,89,101,105,121,130,143,162,190,195,232,267,270,340,352,381,397,417,504,513,546,570 'workbook-loc':104,194,339 'workbook-vis':129,142,161,351,396,416,503,512,545 'workflow':535,581 'worksheet':285,456,458,466 'world':586 'write':290 'writeback':501 'yet':251","prices":[{"id":"c9c0d926-9319-4ebd-82a7-25c9968be44b","listingId":"d483a8f6-cedf-42e7-9b7e-ffc3fe66b2b9","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"dream-num","category":"skills","install_from":"skills.sh"},"createdAt":"2026-04-23T13:03:52.146Z"}],"sources":[{"listingId":"d483a8f6-cedf-42e7-9b7e-ffc3fe66b2b9","source":"github","sourceId":"dream-num/skills/agent-sheet","sourceUrl":"https://github.com/dream-num/skills/tree/main/skills/agent-sheet","isPrimary":false,"firstSeenAt":"2026-04-23T13:03:52.146Z","lastSeenAt":"2026-05-14T01:04:28.573Z"}],"details":{"listingId":"d483a8f6-cedf-42e7-9b7e-ffc3fe66b2b9","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"dream-num","slug":"agent-sheet","github":{"repo":"dream-num/skills","stars":14,"topics":["agent-skills","claude-code-skills","claude-skills","cli","openai-codex","openclaw-skills","skills","spreadsheet"],"license":"apache-2.0","html_url":"https://github.com/dream-num/skills","pushed_at":"2026-05-13T08:31:00Z","description":"Official Univer skills for cli, shell-native spreadsheet automation.","skill_md_sha":"e1a7fd490450526cb591bbfd6662dfb4eb4bffce","skill_md_path":"skills/agent-sheet/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/dream-num/skills/tree/main/skills/agent-sheet"},"layout":"multi","source":"github","category":"skills","frontmatter":{"name":"agent-sheet","description":"Deprecated legacy alias for `univer-cli`; use only for older prompts or installs that still call `agent-sheet`. Prefer `univer-cli` for current shell-native workbook work."},"skills_sh_url":"https://skills.sh/dream-num/skills/agent-sheet"},"updatedAt":"2026-05-14T01:04:28.573Z"}}