Skip to content
This repository was archived by the owner on Sep 23, 2025. It is now read-only.

Commit 8c51ae2

Browse files
committed
update intro
1 parent 0402ae1 commit 8c51ae2

File tree

1 file changed

+19
-2
lines changed

1 file changed

+19
-2
lines changed

src/introduction.md

Lines changed: 19 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,12 +2,29 @@
22

33
This repository explores techniques for making use of Claude Code, Q CLI, and other similar AI assistants. The part of this repository that is currently actionable is the collection of prompts that I have found useful. These include [prompts meant to be installed user-wide][user-prompts] and add-on prompts associated with [memory retention approaches][memory-approaches].
44

5+
## TL;DR: How do I install the damn things
6+
7+
To install the user-wide prompt, checkout the repo
8+
9+
```bash
10+
cd /path/to/socratic-shell
11+
./src/prompts/user/install.sh
12+
```
13+
14+
The optional per-project prompts are meant to be sync'd manually and stored within each project. That way you are working with a snapshot that only changes when you want it to. You can use this script to do that. Then you add things like `@.socratic-shell/README.md` to your project's CLAUDE.md.
15+
16+
```bash
17+
curl https://raw.githubusercontent.com/socratic-shell/socratic-shell/main/src/prompts/project/install.sh | bash
18+
```
19+
20+
# Team Collaboration Patterns
21+
22+
```
23+
524
[user-prompts]: https://github.com/nikomatsakis/socratic-shell/tree/main/prompts/user
625
726
[memory-approaches]: ./retaining-context.md
827
9-
The second part of this repository is source code towards an experimental memory system for retaining context across sessions more automatically. The memory system is designed as an MCP tool that integrates with the above prompts, using "hooks" to trigger memory operations.
10-
1128
## The goal: AI as a pair programming partner
1229
1330
Most AI tools seem to be geared for action -- they seem to be designed to wow you by creating functional code from minimal prompts. That makes for an impressive demo, but it doesn't scale to real code. What I and others have found is that the best way to work with AI assistants is to use them as your **pair programming partner**. That is, talk out your designs. Sketch. Play. Work top-down, just as you would with a human, avoiding the need to get into details until you've got the big picture settled. *Then* start to write code. And when you do, *review*

0 commit comments

Comments
 (0)