원클릭으로 Manus에서 모든 스킬 실행

lumi-reading-chapter-ingest

스타34

포크14

업데이트2026년 7월 7일 14:55

Ingest one book chapter into the wiki and extract character mentions, theme tags, and plot beats into the graph. Use whenever the user wants to ingest, file, or process a chapter — including phrasings like "I just finished chapter 3", "add this chapter to the wiki", "file chapter 2 of [book]", or "track what happened in chapter 5".

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

tronghieu

tronghieu/lumina-wiki

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

관련 직업SOC

SOC 직업 분류 기준

기술 작가예술, 디자인, 엔터테인먼트, 스포츠, 미디어직·SOC 27-3042

SKILL.md

readonly

이 저장소의 다른 Skills

같은 저장소

lumi-check

tronghieu/lumina-wiki

Run lint.mjs --json, summarize findings by severity, offer to apply --fix for auto-fixable checks (L01/L03/L06/L07/L09), self-check re-run to confirm 0 errors, and surface advisory warnings for user attention. Use this whenever the user asks to "check the wiki", "run lint", "verify the graph", "are there broken links?", "what's wrong with the wiki?", "health check", or "are there missing reverse links?". Also fires for: weekly review requests, before committing wiki changes, after a bulk ingest session, or any time the user wants to know the structural state of the wiki.

2026-07-1834

lumi-edit

tronghieu/lumina-wiki

Guided wiki page authoring and refactoring: apply user-directed edits to a wiki page with schema validation and bidirectional-edge integrity preserved. Use this whenever the user asks to update, fix, revise, rewrite, correct, or clean up a wiki page — even if they don't say "edit". Also fires for: "fix the reverse link on X", "update the summary for Y", "set importance to 4 on Z", "the lint check found a missing link in ...", or any request that modifies one wiki page without ingesting a new source.

2026-07-1834

lumi-ingest

tronghieu/lumina-wiki

Turn a raw source file into structured wiki pages: source page, concept and person stubs, wiki.mjs-managed graph edges, citation edges, and a log entry. Pauses at four checkpoints (draft, lint, verify, finalize) so you can review before the work is committed to the wiki — accept, revise, or quit at any checkpoint without losing progress. Resumable across sessions. Use this whenever the user says "ingest", "add", "file", "process", "summarize into the wiki", "create a wiki page for", or drops a filename from raw/sources/. Also fires for: "I added a PDF to raw/sources/", "add this paper to the wiki", "parse this article", "what should I do with raw/sources/X?", or any request to bring a new source into the wiki graph. Also accepts Mode B input — paper title, arxiv ID, or URL, without a local file path. Examples: "ingest paper 2604.03501v2", "ingest arxiv:2604.03501", "ingest https://arxiv.org/abs/2604.03501". The skill fetches the PDF automatically in that case. Also handles whole books, theses, and other 50+ p

2026-07-0734

lumi-verify

tronghieu/lumina-wiki

Check that wiki notes match the sources they cite. Flags statements that appear in a wiki page but are not supported by the cited material, statements that contradict other parts of the same page, and (with --external) statements contradicted by current web search results. Reports findings for the user to review; never edits notes automatically. Run on one page (`/lumi-verify <slug>`), the whole wiki (`/lumi-verify --all`), or only recently-changed pages (`--since <date>`). Use after AI adds new wiki pages, before sharing or exporting the wiki, or as a periodic health check on older pages. Trigger on phrases like "verify", "fact-check the wiki", "is this page accurate?", "check sources", "retro verify". Operates on sources/* pages in v0.9; other page types are skipped.

2026-07-0734

lumi-ask

tronghieu/lumina-wiki

Answer questions from what the wiki already knows, citing the wiki pages behind every claim so the user can open and verify them. Reads only — never changes the wiki unless the user explicitly asks to save the answer as a page. Use this whenever the user asks a question whose answer should come from the wiki — even if they don't say "ask". Also fires for: "what does the wiki say about X", "compare X and Y from what we've ingested", "summarize the wiki's coverage of X", "what concepts relate to Y", "find all sources that mention Z", "what have we learned about X", or any synthesis or retrieval request over accumulated wiki content. If a question could be answered from raw/ but isn't in the wiki yet, surface the gap, list the matching raw/ files so the user can open them directly, and suggest /lumi-ingest instead of reading raw/.

2026-07-0534

lumi-init

tronghieu/lumina-wiki

Bootstrap a fresh Lumina wiki workspace: verify config, materialize directory structure, seed wiki/index.md and wiki/log.md, and write the first log entry. Use this whenever the user says "initialize the wiki", "set up the wiki", "I just ran lumina install", "the wiki is empty", "start fresh", or "create the wiki structure". Also fires for: "wiki/index.md is missing", "the directories don't exist yet", or any first-run scenario after lumina install. Run once per workspace. Safe to re-run (idempotent).

2026-07-0234

name	lumi-reading-chapter-ingest
description	Ingest one book chapter into the wiki and extract character mentions, theme tags, and plot beats into the graph. Use whenever the user wants to ingest, file, or process a chapter — including phrasings like "I just finished chapter 3", "add this chapter to the wiki", "file chapter 2 of [book]", or "track what happened in chapter 5".
allowed-tools	["Bash","Read","Write","Edit"]

/lumi-reading-chapter-ingest

TL;DR

Turns one chapter (raw text or PDF page range) into a wiki/chapters/<book-slug>/ page, seeds character stubs, records theme tags, writes plot beats, and registers bidirectional edges — all idempotently. Re-running against the same slug is always safe.

All graph and frontmatter mutations go through _lumina/scripts/wiki.mjs. Never edit wiki/graph/edges.jsonl directly. add-edge is idempotent and writes reverse edges automatically except terminal/symmetric cases.

When to use

Invoke when the user hands you a book chapter — either as raw text pasted inline, a file path under raw/, or a PDF with a page range (e.g. raw/sources/gatsby.pdf 12-34).

This skill follows the reader chapter by chapter, so the wiki never knows more of the book than the user does — that is what keeps /lumi-reading-plot-recap spoiler-safe. If the user instead wants a finished book absorbed whole ("ingest this book", "add the whole book to the wiki"), route to /lumi-ingest — its long-source pipeline reads the entire document and writes page-anchored reading notes. Do not run this skill dozens of times as a substitute, and do not let /lumi-ingest deep-read a novel the user is still reading. This skill writes the chapter page, seeds character stubs, records theme tags, and registers plot beats. It is designed to be run once per chapter and is fully idempotent: re-running against the same chapter slug produces byte-identical output.

Inputs

<book-slug> — kebab-case identifier for the book. All reading-pack pages for this book live under <book-slug>/ subdirectories. Namespacing is mandatory: a workspace may hold multiple books and the same character name must not collide across books.
<chapter-reference> — one of:
- Raw text (pasted or piped)
- A file path: raw/sources/<book-slug>/<filename>
- A PDF page range: raw/sources/<book-slug>.pdf <start>-<end>
<chapter-number> — integer (1-based). Stored as number: in frontmatter; used by plot-recap for spoiler-boundary ordering.
<chapter-title> — human-readable title (converted to slug automatically).

Workflow

Playbook A: Raw text input

Receive text. Run node _lumina/scripts/wiki.mjs slug "<chapter-title>" --json to get the canonical slug. Chapter page path: wiki/chapters/<book-slug>/<chapter-slug>.md.
Check idempotency: if the page already exists, read its frontmatter with node _lumina/scripts/wiki.mjs read-meta chapters/<book-slug>/<chapter-slug> --json. If number, book, and title match the current inputs, skip page creation and proceed directly to the edge-sync step (step 5). This ensures re-runs are no-ops.
If the chapter page is new, write the full markdown file with valid frontmatter and body via Write. For an existing page, update frontmatter with set-meta. Required fields: id, title, type: chapter, created, updated, book, number.
Extract from the chapter text:
- Character mentions: proper nouns that appear as actors in the narrative. Create a stub entry for each new character (see character-track skill for full character pages).
- Theme tags: 2-5 thematic labels (e.g. isolation, class-conflict, memory).
- Plot beats: 3-7 one-sentence event summaries in narrative order.
Register edges using node _lumina/scripts/wiki.mjs add-edge:
- For each character mentioned: add-edge chapters/<book-slug>/<chapter-slug> features characters/<book-slug>/<character-slug>
- The engine writes the appears_in reverse edge automatically.
- For each theme: add-edge chapters/<book-slug>/<chapter-slug> tagged_with themes/<book-slug>/<theme-slug>
- The engine writes the appears_in reverse edge automatically.
Write plot beats into wiki/plot/<book-slug>/ch<N>-beats.md with valid plot frontmatter (book, up_to_chapter: <N>, etc.) and body.
Update wiki/index.md so the new chapter, character/theme stubs, and plot page are cataloged. Prefer the workspace linter/index workflow when available; do not edit graph files by hand.
Append one line through the wiki engine:
```
node _lumina/scripts/wiki.mjs log reading-chapter-ingest "<book-slug> ch<N> \"<chapter-title>\" -> <K> characters, <M> themes"
```
Log entry text: ## [YYYY-MM-DD] reading-chapter-ingest | <book-slug> ch<N> "<chapter-title>" → <K> characters, <M> themes
Self-verification: run node _lumina/scripts/wiki.mjs read-edges chapters/<book-slug>/<chapter-slug> --json and confirm that at least one features edge and one tagged_with edge are present. If either is missing, add the missing edges before finishing.
Run node _lumina/scripts/lint.mjs --json when the workspace has the linter. Use --fix only for index/frontmatter fixes that match the user's requested scope.

Playbook B: PDF page range

Follow Playbook A steps, but in step 1 extract text using the bundled tool:

python3 _lumina/tools/extract_pdf.py raw/sources/<book-slug>.pdf --pages <start>-<end>

If the tool exits 3 with pip install pypdf, ask the user to run that and retry. If it warns "may be scanned", inform the user and ask them to paste the text directly. Claude Code's native Read tool also parses PDFs and is an acceptable substitute when running in Claude Code.

Output / DoD

wiki/chapters/<book-slug>/<chapter-slug>.md exists with valid frontmatter (id, title, type, created, updated, book, number).
wiki/characters/<book-slug>/<character-slug>.md stub exists for each mentioned character.
wiki/themes/<book-slug>/<theme-slug>.md stub exists for each tagged theme.
wiki/plot/<book-slug>/ch<N>-beats.md exists with the chapter's plot beats.
Bidirectional edges registered: features/appears_in and tagged_with/appears_in.
wiki/index.md updated. wiki/log.md appended.
Lint/check run where available; unresolved issues are reported with exact slugs.
Re-running this skill against the same chapter slug produces byte-identical files.

Guardrails

Use Write only when creating a new page with complete frontmatter. For frontmatter mutations on an existing page, use wiki.mjs set-meta.
Graph/frontmatter mutation must go through _lumina/scripts/wiki.mjs; never edit wiki/graph/edges.jsonl, generated citation files, or existing frontmatter by raw text edits.
Do not pass --book to add-edge; the <book-slug> namespace is part of the slug path, e.g. characters/<book-slug>/<character-slug>.
Do not add reverse edges manually for features or tagged_with; add-edge writes appears_in automatically and remains a no-op on re-run.
Do not infer character genders, ages, or relationships from a single chapter. Record only what the text states directly.
Limit theme tags to 2-5 per chapter. Over-tagging dilutes theme pages.
The number frontmatter field on the chapter page is the authoritative ordering key for plot-recap. Set it correctly; do not guess.

Examples

Input: user pastes text of chapter 1 of "The Great Gatsby", book-slug `great-gatsby`, chapter-number 1, title "In My Younger and More Vulnerable Years". Action: slug -> `in-my-younger-and-more-vulnerable-years`. Writes: `wiki/chapters/great-gatsby/in-my-younger-and-more-vulnerable-years.md` with `number: 1`. Extracts characters Nick, Gatsby, Daisy, Tom, Jordan. Themes: `class`, `the-american-dream`. Registers 5 `features` edges + 5 reverse `appears_in` edges. Writes `wiki/plot/great-gatsby/ch1-beats.md`. Input: same chapter, re-run. Action: read-meta returns matching number/book/title -> skip page creation, run edge-sync only -> add-edge calls return no-op (idempotent) -> byte-identical output confirmed. Input: chapter 2 of the same book, introduces new character "Myrtle". Action: creates `wiki/characters/great-gatsby/myrtle.md` stub (first_seen: ch2). Adds `features`/`appears_in` edges. Existing character pages (Nick, Gatsby) are NOT rewritten by this skill; character-track handles updates. Input: `raw/sources/gatsby.pdf 35-67`, chapter 3, title "The Parties Begin". Action: reads pages 35-67, follows Playbook B. If text is extractable, proceeds normally. If scanned, responds: "This PDF appears to be scanned. Please paste the chapter text."