# Map

> Show literature-to-chapter coverage and the writing progress dashboard — which sources support which chapters, and word counts against targets.

- Skill: `yha9806/map` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add yha9806/map`
- Raw SKILL.md: https://api.skillmd.com/api/skills/yha9806/map/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing
- Author: yha9806 (https://skillmd.com/u/yha9806)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/yha9806/map

---


# /map — Literature-Thesis Mapping Skill

## Purpose

Build and display a mapping matrix showing which literature sources support which thesis chapters. Identify coverage gaps where chapters lack sufficient source support.

## Trigger Words

This skill activates on: `map`, `show mapping`, `coverage`, `which sources`, `/map`.

## Workflow

1. **Scan all notes files** in `literature/reading_notes/` using Glob (`*_NOTES.md`). For each file, extract:
   - Author and title (from the `# Reading Notes:` heading)
   - `Status` field value
   - `Relevance` field value
   - All rows from the `## Thesis Connections` table

2. **Scan chapter files** in `chapters/` to get the list of chapters and their section headings.

3. **Build the mapping matrix.** Rows are sources (sorted alphabetically by author), columns are chapters (Ch1 through Ch8 or however many exist).
   - Mark each cell with the connection type if a mapping exists (e.g., `supports`, `challenges`, `extends`).
   - Leave cells empty if no connection.

4. **Identify coverage gaps.** Flag any chapter that has fewer than 3 mapped sources as "under-covered".

5. **Add the progress dashboard.** Count chapter words deterministically with
   `node .claude/skills/map/scripts/count-words.mjs --base-dir . --json` (never estimate word counts by reading), compare
   against the per-chapter targets in `CLAUDE.md`, and count notes files by
   `Status` value (reading / completed / integrated).

6. **Output the mapping and dashboard.**

## Output Format

```
## Literature-Thesis Mapping -- {YYYY-MM-DD}

### Matrix

| Source | Status | Ch1 | Ch2 | Ch3 | Ch4 | Ch5 | Ch6 | Ch7 | Ch8 |
|--------|--------|-----|-----|-----|-----|-----|-----|-----|-----|
| Smith (2024) | integrated | | supports | supports | | | | | |
| Jones (2021) | integrated | extends | supports | supports | extends | | | | |
| Lee (2019) | completed | | | supports | | | challenges | | challenges |

### Coverage Summary

| Chapter | Mapped Sources | Status |
|---------|---------------|--------|
| Ch1 | {N} | OK / Under-covered |
| Ch2 | {N} | OK / Under-covered |

### Under-covered Chapters

{List chapters with < 3 sources, with suggestions for which completed-but-not-yet-mapped notes might fill the gap.}
```

## Saving the Matrix

If the user says "save" or "export mapping", write the matrix to `literature/mapping_matrix.md` using Edit (if the file exists) or Write (if creating new). Include a timestamp header.

## Gap Threshold

- **Under-covered**: fewer than 3 mapped sources per chapter.
- This threshold can be adjusted if the user specifies a different minimum.

## Constraints

1. **Read-only by default.** Only modify files if the user explicitly asks to save the matrix.
2. **No emoji** in output.
3. **No hardcoded chapter count.** Detect chapters dynamically from the `chapters/` directory.
4. **Source names** use the format `{Author} ({Year})` for readability.
5. **Connection types** are derived from the `Connection Type` column in notes files. If a notes file lacks the table or fails the contract lint (`npm --prefix guards run lint:notes -- {file}` when `guards/` is present), list it under "Unparseable notes files" instead of inferring cells — a broken file must surface as an error, never as a guessed mapping.
6. **Word counts** come from the bundled counter, never from model estimation. It counts whitespace-delimited words, including Markdown tokens, as the former `wc -w` instruction did; continuous Chinese text is not counted character by character.