Home / All Skills / Workflow & Core Tools / ljg-learn
Workflow & Core Tools · lijigang/ljg-skills

ljg-learn

Deep concept anatomist that deconstructs any concept through 8 exploration dimensions (history, dialectics, phenomenology, linguistics, formalization, existentialism, aesthetics, meta-philosophy) and compresses insights into an epiphany. Use when user asks to explain, dissect, or deeply understand a concept, term, or idea. Triggers on '解剖概念', '概念解剖', 'explain concept', 'learn concept', '/ljg-learn'. Produces org-mode output.

L2 Local · Restricted workspaceV1 Commit-PinnedWrite FileDual-pass Audit
Author lijigangGitHub lijigang/ljg-skills ↗License MIT(仓库根 LICENSE)Commit 9e75497a95
Control levels describe skill capabilities and suggested control measures; they do not represent quality or safety certifications, nor do they dictate usability. Usage decisions rest entirely with users and administrators. Different agent hosts apply varying constraints; do not assume hosts provide automatic protection. Third-party skill copyrights belong to their original authors. Use at your own risk.

1How It Works & Mechanism

Overall Stance: Treat any concept as an anatomical subject, dissect it across eight dimensions, and finally compress it into an epiphany.

skills/ljg-learn/SKILL.md
你是概念解剖师。拿到一个概念,从八个方向切开它,最后把所有切面压成一句顿悟。
What this does: Establishes the skill's role and guiding principle—eight-dimensional slicing + final compression, rather than an encyclopedic explanation.

Step 1 'Anchoring': Clarify common definitions and misconceptions first, then unpack core morphemes/etymological roots.

skills/ljg-learn/SKILL.md
1. 这个概念最通行的定义是什么?常见误解在哪?
What this does: Fixes the semantic boundaries of the subject before making the eight cuts, preventing deviations.

Step 2 'Eight Cuts': Make one cut across each of the eight dimensions, keeping only the skeletal essence in 2-3 sentences per cut.

skills/ljg-learn/SKILL.md
八个方向各切一刀。每刀 2-3 句,只留筋骨,不带水分。
What this does: Uses fixed eight dimensions as an anatomical scaffold and enforces length limits to drive density.

The eight dimensions consist of History, Dialectics, Phenomenology, Linguistics, Formalization, Existentialism, Aesthetics, and Meta-reflection, each with defined action items.

skills/ljg-learn/SKILL.md
**历史**:最早从哪冒出来 → 怎么变的 → 哪一步拐成了今天的意思
What this does: Translates the abstract 'eight cuts' into eight actionable, concrete prompts.

Step 3 'Introspection': Embody the concept itself to view the world in the first person, identifying cuts that point to the same deep structure.

skills/ljg-learn/SKILL.md
变成这个概念本身,用第一人称看世界。3-5 句。
What this does: Restores the experience from inside the concept and converges the eight cuts onto a shared deep structure.

Step 4 'Compression': Generate a three-piece set consisting of a formula, a one-sentence epiphany, and a pure ASCII structural diagram.

skills/ljg-learn/SKILL.md
**结构图**:纯 ASCII 画出概念的骨架
What this does: Compresses the anatomical findings into a minimal, repeatable, and drawable form.

Formatting Iron Rule (Zero Exceptions): Output must be pure org-mode; Markdown syntax is explicitly prohibited.

skills/ljg-learn/SKILL.md
输出必须是纯 org-mode 语法,禁止任何 markdown 语法
What this does: Enforces output syntax with a zero-exception rule to ensure seamless integration into org-mode note systems.

Write Workflow: Retrieve timestamp, write to ~/Documents/notes/, and report the full path.

skills/ljg-learn/SKILL.md
运行 `date +%Y%m%dT%H%M%S` 获取时间戳。
What this does: Saves the final output as a timestamped org file and returns the path via local host actions.

2Core Capabilities

01Eight-dimensional concept dissection: one cut each across history / dialectics / phenomenology / linguistics / formalization / existentialism / aesthetics / meta-reflection
02Anchoring: clarify common definitions, dispel misconceptions, and unpack core morphemes
03Introspection: reconstruct the concept's experience in first-person and extract shared deep structures
04Compression trio: formula + one-sentence insight + pure ASCII structural diagram
05Pure org-mode output, strictly adhering to zero-exception formatting rules (Markdown syntax prohibited)
06Automatic naming and writing to local ~/Documents/notes/ knowledge base
07Chinese trigger phrases and slash command /ljg-learn entry point

4Recommended Governance & Evidence

Suggested Controls for Safe Operation
For Agent Runtime
Restrict working directory
For Administrators
Retain operation logs
Evidence & Triggers (File Paths & Excerpts)
  • 本地工作目录: File I/O and artifact output: ~/Documents/notes/{timestamp}--概念解剖-{概念名}__concept.org (written via host Write tool, skill
Audit Notes: Pure prompt-based workflow: Directory contains only one SKILL.md, with no scripts, network requests, or credential reads. The body instructs the host to run date to get a timestamp and write a local org file, which is a local host Bash/Write action; based on assessment criteria, the rating tier remains unchanged, recorded only under file_writes.

5Second-Pass Verification

  • [ok] 八维清单完整 — §八刀列出历史/辩证/现象/语言/形式/存在/美感/元反思并各附动作,与引用一致。
  • [ok] 五步流程结构 — §1-§5(定锚/八刀/内观/压缩/写入)齐备,压缩内含公式、一句话、结构图三项。
  • [ok] 格式铁律五条 — §写入列出纯 org-mode、粗体 *bold*、禁用 ---、列表语法、~code~ 五条,与 traits 一致。
  • [ok] 写入路径与命名 — 写 ~/Documents/notes/{timestamp}--概念解剖-{概念名}__concept.org,时间戳由 date 取得,逐字命中。
  • [discrepancy] 扫描报无外部程序/写盘 vs 正文要求 date 与写文件 — 风险扫描六类均零命中,但 §写入明确要求运行 date 并写 org 文件;判为宿主本地动作(非 skill 自带脚本),已记入 file_writes 并在 reason 说明。
  • [ok] 外部依赖与凭证 — 全文无 URL、无 API、无包依赖,external_deps 据实留空。

6Verdict & Scenarios

  • Structured eight-dimensional dissection scaffolding that prevents concept explanations from devolving into generic platitudes.
  • Strict formatting discipline (pure org-mode, ASCII diagrams) allowing direct integration into org note-taking workflows.
  • Compression into 'formula + one sentence + structural diagram' forces out repeatable core insights.
  • Best for: Ideal for writers and learners seeking to deeply dissect a concept quickly and structurally, especially users who preserve thoughts in org-mode notes; also suitable as a templated service for thoroughly mastering a term.
    Not for: Not suitable for: conceptual research requiring authoritative citations and references; users seeking encyclopedic comprehensive definitions; non-org-mode workflows without local file writing capabilities; and scenarios with zero tolerance for non-factual interpretations that reject philosophical analysis.
    Install Agent-Ready
    ① Verified Mirror 2026-10-10
    Option A · Direct DownloadDownload ljg-learn.tar.gz
    sha256: 950eda1d911d5e32…
    Option B · Copy Guide for Agent
    install.json
    Agents automatically download and verify mirrors from this JSON guide.
    ② Upstream GitHub
    GitHub Repo ↗
    Pinned to commit 9e75497a95.