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.

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.

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

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.

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".

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/.

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).

Backfill missing schema fields (provenance, confidence) on legacy wiki entries after a Lumina version upgrade. Use proactively when /lumi-check reports L01/L11 findings on multiple entries, when manifest shows legacyMigrationNeeded:true, or when user mentions "lint do", "upgrade", "missing provenance", or "schema gap" after running install.

Scoped destructive reset with confirmation UX: show a dry-run plan, require explicit user confirmation in chat, then execute reset.mjs. Five scopes: wiki, raw, state, checkpoints, all. Always dry-runs first — never skips the plan step. Use this whenever the user says "reset the wiki", "clear everything", "start over", "delete all wiki pages", "wipe the raw folder", "clear stuck checkpoints", "the ingest got stuck and I want to restart", or any request to bulk-delete wiki or raw content. Also fires for: "I want a clean slate", "remove all ingested content", "clear the state files". When the scope is ambiguous, ask before running dry-run — "did you mean wiki/, raw/, or all (wiki + state)?"

Guide a self-reflection session on a concept, source, or topic the user has studied. Creates or updates a personal reflection page in wiki/reflections/ with a rewritable "Current understanding" section and an append-only "Evolution" log. AI acts as a metacognitive mirror — quotes the user's past words and asks questions — but never writes reflection content. Use whenever the user invokes /lumi-learning-reflect, says "reflect on <concept>", "update my reflection on <topic>", "what do I think about <concept>?", "I want to journal about <topic>", "phản tư về <concept>", "suy ngẫm về <topic>", "反思 <concept>", "我对 <topic> 的理解", or any phrasing indicating they want to record or revisit their personal understanding of a wiki concept or source.

Maintain character pages in wiki/characters/<book-slug>/ and update inter-character relationship edges as new chapters are ingested. Use whenever the user wants to update or review character information — including phrasings like "update character profiles after chapter 4", "add the relationship between Alice and Bob", "who has Alice met so far", "track character appearances", or "refresh character pages".

Build and maintain thematic cluster pages in wiki/themes/<book-slug>/ by linking themes to the chapters and characters they appear in. Use whenever the user wants to analyze, map, or update themes — including phrasings like "map out the themes", "cluster the themes after chapter 5", "what themes appear across the book", "theme analysis", or "update theme pages".

Seed terminal foundation pages for stable background knowledge so future ingestion can link to common concepts without duplicating definitions.

Score an already-ingested paper's influence and quality. Fetches citation influence (and optional Scite/Altmetric signals when keys are set), estimates venue prestige, and runs a structured 4C quality assessment, then writes a transparent ranking block onto the source page. Helps prioritize what to read next.

Produce a grounded survey from existing wiki sources, concepts, and graph edges. Writes a summary page only when the user asks to save the survey.

Cluster existing concepts and sources into a thematic topic page under wiki/topics/ so related knowledge is grouped and cross-referenced in the graph.

Run one scheduled-discovery pass over the consolidated watchlist (topics + feeds). Use this whenever the user asks to "check for new papers", "run discovery now", "poll my feeds", "see what's new this week", or asks about the watchlist findings. Also fires when the user wants to wire a cron / launchd / Task Scheduler job — point them at the wrapper script and docs.

Help the user choose research topics to follow and configure the scheduled discovery watchlist in plain language. Use this whenever the user wants to add, change, pause, remove, or review topics for scheduled discovery, even if they do not mention the watchlist file by name.

Discover candidate sources for a research topic using the opt-in Python research tools, rank them, and present a shortlist for user approval. This skill proposes sources; it does not ingest automatically.

Configure the research pack runtime by checking Python dependencies and helping the user populate local API-key environment files.

Orient the user in their Lumina wiki workspace. Three modes: Orientation (default — recommend ONE next action; offer to run), Catalog (on `skills`/`list` arg or features question — render lumi-help.csv grouped by pack), Framework Q&A (on `explain` arg or how-it-works question — answer from local docs with citations). Use when the user says "help", "what next", "I'm lost", asks for orientation, or asks how Lumina works.

Generate a spoiler-safe progressive recap of a book up to (but NOT including) the user's current chapter cursor. Use whenever the user asks for a summary of what has happened so far — including phrasings like "where am I in the story", "what's happened so far", "recap up to chapter 3", "remind me of the plot", "catch me up", "what do I know so far", or "summarize the story to this point".

Create high-quality, model-aware system prompts for any LLM (Claude, GPT, Gemini, open-source, etc.). Use this skill whenever the user wants to create, write, build, design, draft, or improve a system prompt, system instructions, or custom instructions for any AI model. Also trigger when the user asks about prompt engineering, prompt design, prompt optimization, or wants to define behavior for an AI assistant, chatbot, agent, or any LLM-powered application — even if they don't explicitly say "system prompt". Covers all use cases including chatbots, agentic systems, tool-use workflows, content generation, data extraction, code assistants, and multi-turn conversations.

Profile and optimize Python code using cProfile, memory profilers, and performance best practices. Use when debugging slow Python code, optimizing bottlenecks, or improving application performance.