菜单
Guide a new or returning user through loom setup, configuration, first sync, history intake, and the code-vs-data boundary. Use when the user asks to initialize loom, follow ONBOARDING.md, configure sources, collect historical work traces, or explain what lives in the repo versus ~/.loom.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
确保 loom CLI 已安装并在 PATH 上。当准备使用 loom 但命令行里 `loom` 不存在(command not found / 首次使用 loom 插件)时,先用本技能一行装好底层 CLI,再继续。仅负责安装引导,不负责日常使用(用法见 loom 技能)。
同步/检索个人跨工具成果台账 loom。当用户想「记录今天做了什么」「同步 loom」「查我以前/某天/某件事做过什么、结论是什么、相关文档/数据在哪」「生成日报/周报/绩效素材」时使用;先用它查历史,别只翻当前仓。
Generate and write back AI-authored loom daily reports from real collected evidence. Use when the user asks for a daily report, work log, day summary, report gen/set workflow, or wants to sync loom, draft 今日工作/今日思考/明日计划, and store the result.
Generate accurate loom AI session digest TSVs from real question-and-answer transcripts and write them back. Use when the user asks to summarize sessions, improve vague AI conversation titles, run loom session gen/set/ls, batch process Claude sessions, or make answer-side content searchable.
Classify loom entries into topic leaves with a gather-to-TSV-to-apply workflow and post-apply verification. Use when the user asks to triage topics, build or refine the topic DAG, run loom topic gather/apply/show, review topic assignments, or classify historical entries without keyword overreach.
Inspect loom vault backup safety before git push or cloud backup. Use when the user asks to check vault state, gitignore coverage, remote configuration, tracked raw data, push risk, private backup readiness, or whether loom sync --push is safe.
| name | loom-onboarding |
| description | Guide a new or returning user through loom setup, configuration, first sync, history intake, and the code-vs-data boundary. Use when the user asks to initialize loom, follow ONBOARDING.md, configure sources, collect historical work traces, or explain what lives in the repo versus ~/.loom. |
Use ONBOARDING.md as the source-of-truth runbook. Work one phase at a time, inspect command output before proceeding, and do not skip the privacy boundaries.
If loom is not on PATH, run commands from the repo root with python3 -m loom.
--push, creating a remote repo, sharing data, or writing cron/automation.~/.loom/.env with mode 600. Never paste secrets into repo files, markdown notes, logs, or chat output.~/.loom/ is the private instance containing config.json, .env, data/, and vault/.loom sync as local collection/render/vault commit. Treat loom sync --push as external backup and require confirmation.Run:
loom --help || python3 -m loom --help
python3 -m pytest tests/ -q
Ask for the user's name, git author emails or names, and the root directory to scan for git repos. Default the scan root to ~/Documents if the user has no preference.
Run:
loom init
Let the user enter secrets interactively when Feishu is configured.
Run without push:
loom sync
loom today
loom search <keyword>
Explain the result in concrete terms: entries live in ~/.loom/data/entries.jsonl, search index and journals are derived, and collection is read-only except for writing the loom instance.
Use explicit commands for loose material:
loom doc add <paths...> --to <category> --tags <tags>
loom data add <csv-or-xlsx...> --to <topic> --from <upstream> --code <code-files>
loom note "<text>" --to <category> --tags <tags>
loom report import <daily-report.xlsx>
Do not bulk-import raw data by directory. CSV/XLSX files must go through loom data add so raw files remain local under _data/ and only data cards become searchable.
Only after the user confirms private backup, verify the vault remote and risk with $loom-vault-ops. Then run:
loom sync --push
Do not set cron or recurring automation without explicit confirmation.
For topic classification, switch to $loom-topic-triage. For daily reports, use $loom-daily-report. For session summaries, use $loom-session-digest.