second-brain-report

1# Write a research report
2 
3A report that presents only findings looks stronger and is less useful, because
4the reader cannot tell where it is load-bearing and where it is thin.
5 
6## Core rule
7 
8Scope against the vault first. Most of the work in a good report is knowing what
9you do not need to look up.
10 
11## Workflow
12 
131. **Inventory what exists.** Which pages cover this, what is thin, what is
14   missing entirely. That answer is the research plan.
152. **Report the gaps before writing,** so the user can decide whether to ingest
16   more first.
173. **Write from pages, citing pages.** Every claim traces to a source page with
18   a URL behind it.
194. **Record disagreement** where sources conflict, with dates and who says what.
205. **State what could not be established** and what evidence would change the
21   conclusion.
226. **Save to `output/`** and offer to ingest it back as a source.
23 
24## Output format
25 
26```
27## Question
28## What the vault covers    (with [[page]] citations)
29## Where sources disagree
30## What I could not establish
31## What would change this
32## Pages read: <list>
33```
34 
35## Calibration
36 
37If the vault covers less than half the question, say so at the top rather than
38producing a report that reads as complete.
39 
40Never fill a gap from general knowledge without labelling it as outside the
41vault. A report whose sources cannot be checked is worth less than a short one
42that can.