second-brain-project

1# Set up a project
2 
3The knowledge layer answers what you know. The project layer answers what you
4are doing this week. Merging them fills a wiki with dead task lists and buries
5working files under conceptual notes.
6 
7## Core rule
8 
9One goal per project, stated in its own `CLAUDE.md`. A project with three goals
10is three projects.
11 
12## Workflow
13 
141. **Interview if needed.** What this produces, for whom, by when, and what the
15   agent's role is. Vague answers produce a file that makes every future session
16   worse.
172. **Create the folders:** `Inputs/`, `Process/`, `Outputs/`, `Feedback/`.
183. **Write the project `CLAUDE.md`:** what this is, the one goal with a date,
19   your role, working rules, current state. Name what the agent should not do
20   here as well.
214. **Link the knowledge layer.** Find concept pages the project depends on and
22   reference them from the project file. Copy into `Inputs/` only the few that
23   genuinely matter.
245. **Register it** in the root `CLAUDE.md` project list.
25 
26## Output format
27 
28```
29Project: <name>
30Goal: <one measurable outcome with a date>
31Folders created: Inputs, Process, Outputs, Feedback
32Concept pages linked: <list>
33Agent role: <one line>
34```
35 
36## Calibration
37 
38`Feedback/` is the folder people skip and the one that makes the system improve.
39If the user has no idea what would go in it, that is worth surfacing: it usually
40means the goal is not measurable yet.
41 
42Do not create a project for work that finishes this week. That is a task.