メニュー
Externalize verified quotes and observations to a scratch file as you read, so they survive context compaction. Use during any study session to build a durable working document alongside source reading. This skill exists because files are durable — context is not.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | quote-log |
| description | Externalize verified quotes and observations to a scratch file as you read, so they survive context compaction. Use during any study session to build a durable working document alongside source reading. This skill exists because files are durable — context is not. |
Context windows compact. When a study requires reading 15-25 source files, early readings get compressed into summaries during compaction — losing exact wording, verse numbers, and the small observations that make studies rich. Then the agent re-reads to verify what it already read, burning context on redundancy instead of depth.
The fix: Write verified quotes to a scratch file immediately after reading each source. The file persists across compactions. When it's time to write the study, read the scratch file — not 20 chapters again.
Create a working file at study/.scratch/{topic}.md at the start of the study, right after the outline.
# {Topic} — Source Log
*Working document. Kept as research provenance — traces how observations were reached.*
---
## Outline
1. [Section heading from Phase 1]
2. [Section heading]
3. ...
---
## Verified Quotes
### [Source Path] — [Book Chapter:Verses]
> "Exact quote from Read tool" (v. X)
**Observation:** [What stood out, how it connects, what it might mean]
**Connects to:** [Outline section number or other source]
---
### [Next Source]
> "Quote" (v. X)
**Observation:** ...
---
## Webster 1828
### [Word]
- **Definition X:** "[exact definition]"
- **Study relevance:** [how this illuminates the passage]
---
## Conference Talks
### [Speaker, Title, Date] — [file path]
> "Key quote"
**Observation:** ...
---
## Threads to Pull
- [ ] [Question or connection to explore]
- [ ] [Gap in the outline that needs more sourcing]
## Cross-Study Connections
- [study/related-study.md] — [brief note on connection]
Write immediately after reading. Don't batch. Read a chapter → write the quote and observation → move to the next source. This is the whole point.
Copy exact text. The quote in the scratch file should be copy-paste from what Read returned. No tidying, no paraphrasing. Tidying happens in the draft phase.
Note the outline connection. Even a brief "→ Section III" helps during drafting. You're building a map, not just a list.
Track what's missing. The "Threads to Pull" section is where you note gaps. After reading 10 sources, review this section to decide what else to read — don't re-scan the whole outline from memory.
Don't over-collect. You don't need every verse from every chapter. Grab the quotes that matter for the study's questions. If a whole chapter is relevant, note the range and pull 2-3 key verses.
The scratch file is both scaffolding AND audit trail. It shows how the building was constructed.
Verify scripture and talk quotes against actual source files. Enforces read-before-quoting, the cite count rule, and the pre-publish checklist. Load when writing ANY document that cites scriptures, conference talks, transcripts, or other verifiable sources — studies, evaluations, lessons, guides, docs, everything.
Look up which General Conference talks, Journal of Discourses entries, and other sources cite a particular scripture verse using the BYU Scripture Citation Index (scriptures.byu.edu). Use when you want to know who has taught about a verse, how frequently a verse appears in conference, or to discover that a verse has surprisingly little modern commentary.
Stress-test the study's arguments, assumptions, and gaps before finalizing. Asks 'what are we missing?', 'where are we overreaching?', and 'does the text actually say what we think it says?' Use between gap analysis and first draft to sharpen the study.
Telescope into a specific scripture passage or narrow topic. Follow every footnote, trace every cross-reference, read full chapters for context. Produces intermediate findings in a working document. Use for focused, depth-first study on a single text or concept.
Six text-checkable properties for evaluating non-canonical sources (books, articles, transcripts, web content) before citing or recommending them. Loaded passively whenever the agent is asked to assess, summarize, recommend, or quote from a source outside the LDS canon. Names what the agent CAN check (correspondence, coherence, canonical fit, pillars-rhetoric, mark-keeping, steelman) and what only the user with the Spirit can check (witness, soul-fruit, animating spirit, personal stewardship-fit, override). Grounded in study/discernment-and-the-comprehending-eye.md.
Run the publish script to convert study documents to public HTML, then git add, commit, and push. Use when the user says publish, commit, push, or asks to share their work.