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。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/sancovp/restaurants-module --skill using-restaurants命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | using-restaurants |
| description | Use the restaurants neurosymbolic module: RAG library, agent brain, growable KB — proof-checked |
| version | 0.1.0 |
This module is a CULTIVATED, PROOF-CHECKED knowledge organism about restaurants (2849 concepts / 2429 relations; grown by a KB factory — every region admitted by a Prolog consistency gate, wrongness tracked as open supersede-issues, never hidden).
understand-{x} skills in this plugin,
coordinate-addressed (call number = home class : dependency facets — the
import web, literally); FTS5 index via skilltree.build_index over
${CLAUDE_PLUGIN_ROOT}/skills/using-restaurants/references/skilltree.json.brain_ask("your question"): the activation graph fires
the matching gyri numerically, each answers over its territory, the
synthesis is PROVEN one level up (SES tower) and returns with receipts.ee_v2.kbc.heaven_tools.make_kbc_tools
over this module's data root
(${CLAUDE_PLUGIN_ROOT}/skills/using-restaurants/data/): 14 heaven tools
(kb_, kernel_, brain_*). Hand them to any heaven agent's tools=[...].kb-door issue to point it somewhere; file kb-supersede when
you catch it being wrong (it also catches itself — see the round reports).data/concepts.jsonl + data/relations.jsonl — the certified graphdata/hyperedges.jsonl — the certificate ledger (the automaton's KNOWN)data/skeletons.jsonl — certified argument DAGs (because/since/…)data/worklist.json — what the module knows it doesn't knowreferences/skilltree.json — the library indexkb_work shows exactly what the module knows it
doesn't know.?.american_cuisine?.pos_terminal?.customer_relationship_management_crm?.inventory_management_system?.menu_management_system?.side_dish?.kitchen_display_system?.reservation?.reservation_system?.online_ordering_platform?.analytics_reporting_dashboard?.unified_inventory