4 ms·

I have multiple system prompts that I use before getting to the actual specification. 1. I use the Socratic Coder[1] system prompt to have a back and forth co

I have multiple system prompts that I use before getting to the actual specification. 1. I use the Socratic Coder[1] system prompt to have a back and forth conversation about the idea, which helps me hone the idea and improve it. This conversation forces me to think about several aspects of the idea and how to implement it. 2. I use the Brainstorm Specification[2] user prompt to turn that conversation into a specification. 3. I use the Brainstorm Critique[3] user prompt to critique that specification and find flaws in it which I might have missed. 4. I use a modified version of the Brainstorm Specification user prompt to refine the specification based on the critique and have a final version of the document, which I can either use on my own or feed to something like Claude Code for context. Doing those things improved the quality of the code and work spit out by the LLMs I use by a significant amount, but more importantly, it helped me write much better code on my own because I know have something to guide me, while before I used to go blind. As a bonus, it also helped me decide if an idea was worth it or not; there are times I'm talking with the LLM and it asks me questions I don't feel like answering, which tells me I'm probably not into that idea as much as I initially thought, it was just my ADHD hyper focusing on something. [1]: https://github.com/jamesponddotco/llm-prompts/blob/trunk/data/socratic-coder.md https://github.com/jamesponddotco/llm-prompts/blob/trunk/dat... [2]: https://github.com/jamesponddotco/llm-prompts/blob/trunk/data/brainstorm-specification.md https://github.com/jamesponddotco/llm-prompts/blob/trunk/dat... [3]: https://github.com/jamesponddotco/llm-prompts/blob/trunk/data/brainstorm-critique.md https://github.com/jamesponddotco/llm-prompts/blob/trunk/dat...

time0ut 1y ago
Thank you for sharing these prompts. These are excellent.
indigodaddy 1y ago
Wish we could star files in addition to repos
- addandsubtract 1y ago
  You mean like adding a bookmark, or downloading the files? Yeah, wish that was possible on the web.
  - indigodaddy 1y ago
    Well I use GitHub stars kind of like coding/cool project/idea/whatever bookmarks, so yeah would be neat to be able to star just any file within a repo in addition to the repo itself
ctoth 1y ago
You may want to turn these good prompts into slash commands! :)
- jamesponddotco 1y ago
  They are subagents and slash commands, depending on the project. Eventually, I need to come up with a “dotclaude” repository with these and a few others I use in. Edit: Sorry, I had a brain fart for a second, thought you were talking about other prompts. I prefer to keep those as chats with the API, not Claude Code, but yeah, they might work as slash commands too.
alecco 1y ago
Good stuff. A minor observation: > I use the Socratic Coder[1] system prompt to have a back and forth conversation about the idea. (prompt starts with: 1. Ask only one question at a time) Why only 1? IMHO it's better to write a long prompt explaining yourself as much as possible (exercises your brain and you figure out things), and request as many questions to clarify as possible, review, and suggestions, all at once. This is better because: 1. It makes you think deeper and practice writing clearly. 2. Even though each interaction is quite slower, since you are more active and engaged it feels shorter (try it), and you minimize interactions significantly. 3. It's less wasteful as going back and forth 4. You converge in much shorter time as your misconceptions, misunderstandings, problems expressing yourself, or confusion on the part of the LLM are all addressed very early. 5. I find it annoying to wait for the replies. I guess if you use a fast response conversational system like ChatGPT app it would make more sense. But I don't think that way you can have deep conversations unless you have a stellar working memory. I don't, so it's better for me to write and read, and re-write, and re-read...
- jamesponddotco 1y ago
  I do one question at a time so I don't feel overwhelmed and can answer questions with more details. I start with an idea between <idea> tags, write as much as I possibly can between these tags, and then go one question at a time answering the questions with as much details as I possibly can. Sometimes I'll feed the idea to yet another prompt, Computer Science PhD[1], and use the response as the basis for my conversation with the socratic coder, as the new basis might fill in gaps that I forgot to include initially. [1]: https://github.com/jamesponddotco/llm-prompts/blob/trunk/data/computer-science-phd.md https://github.com/jamesponddotco/llm-prompts/blob/trunk/dat... [2]: Something like "Based on my idea, can you provide your thoughts on how the service should be build, please? Technologies to use, database schema, user roles, permissions, architectural ideas, and so on."
frays 1y ago
Thanks for sharing these prompts. Will certainly help with improving my LLM coding workflow.
philsheard 1y ago
Love these, thanks. What's the benefit of putting the original idea between <idea> tags when it seems to the main body of the prompt anyway? Or are you supplying the Socratic Coder prompt and the idea in the same prompt?
- jamesponddotco 1y ago
  Mostly because the system prompt states "The user will provide the idea you will be working with as the first message between <idea> tags." and Claude loves XML tags, but if you're not including anything else in the message, it probably doesn't matter.
zamber 1y ago
Is this true? I'm getting a feeling that most of this is adding external stucture when coding agents already provide a framework for it. I've had moderate success in throwing a braindump at the llm, asking it to do a .md with a plan and then going with the implementation for it. Specialized thinking prompts seem like overkill (or dumbo-level coding skills are enough for me).
throwup238 1y ago
Heh. I tried your socratic coder prompt as a Claude project and when sending it the Brainstorm Spec message after some back and forth, Claude responded with “Chat ended due to a prompt injection risk”.

time0ut 1y ago

Thank you for sharing these prompts. These are excellent.

indigodaddy 1y ago

Wish we could star files in addition to repos

addandsubtract 1y ago
You mean like adding a bookmark, or downloading the files? Yeah, wish that was possible on the web.
- indigodaddy 1y ago
  Well I use GitHub stars kind of like coding/cool project/idea/whatever bookmarks, so yeah would be neat to be able to star just any file within a repo in addition to the repo itself

ctoth 1y ago

You may want to turn these good prompts into slash commands! :)

jamesponddotco 1y ago
They are subagents and slash commands, depending on the project. Eventually, I need to come up with a “dotclaude” repository with these and a few others I use in. Edit: Sorry, I had a brain fart for a second, thought you were talking about other prompts. I prefer to keep those as chats with the API, not Claude Code, but yeah, they might work as slash commands too.

alecco 1y ago

Good stuff. A minor observation: > I use the Socratic Coder[1] system prompt to have a back and forth conversation about the idea. (prompt starts with: 1. Ask only one question at a time) Why only 1? IMHO it's better to write a long prompt explaining yourself as much as possible (exercises your brain and you figure out things), and request as many questions to clarify as possible, review, and suggestions, all at once. This is better because: 1. It makes you think deeper and practice writing clearly. 2. Even though each interaction is quite slower, since you are more active and engaged it feels shorter (try it), and you minimize interactions significantly. 3. It's less wasteful as going back and forth 4. You converge in much shorter time as your misconceptions, misunderstandings, problems expressing yourself, or confusion on the part of the LLM are all addressed very early. 5. I find it annoying to wait for the replies. I guess if you use a fast response conversational system like ChatGPT app it would make more sense. But I don't think that way you can have deep conversations unless you have a stellar working memory. I don't, so it's better for me to write and read, and re-write, and re-read...

jamesponddotco 1y ago
I do one question at a time so I don't feel overwhelmed and can answer questions with more details. I start with an idea between <idea> tags, write as much as I possibly can between these tags, and then go one question at a time answering the questions with as much details as I possibly can. Sometimes I'll feed the idea to yet another prompt, Computer Science PhD[1], and use the response as the basis for my conversation with the socratic coder, as the new basis might fill in gaps that I forgot to include initially. [1]: https://github.com/jamesponddotco/llm-prompts/blob/trunk/data/computer-science-phd.md https://github.com/jamesponddotco/llm-prompts/blob/trunk/dat... [2]: Something like "Based on my idea, can you provide your thoughts on how the service should be build, please? Technologies to use, database schema, user roles, permissions, architectural ideas, and so on."

frays 1y ago

Thanks for sharing these prompts. Will certainly help with improving my LLM coding workflow.

philsheard 1y ago

Love these, thanks. What's the benefit of putting the original idea between <idea> tags when it seems to the main body of the prompt anyway? Or are you supplying the Socratic Coder prompt and the idea in the same prompt?

jamesponddotco 1y ago
Mostly because the system prompt states "The user will provide the idea you will be working with as the first message between <idea> tags." and Claude loves XML tags, but if you're not including anything else in the message, it probably doesn't matter.

zamber 1y ago

Is this true? I'm getting a feeling that most of this is adding external stucture when coding agents already provide a framework for it. I've had moderate success in throwing a braindump at the llm, asking it to do a .md with a plan and then going with the implementation for it. Specialized thinking prompts seem like overkill (or dumbo-level coding skills are enough for me).

throwup238 1y ago

Heh. I tried your socratic coder prompt as a Claude project and when sending it the Brainstorm Spec message after some back and forth, Claude responded with “Chat ended due to a prompt injection risk”.