HackerReposA FIELD GUIDE TO OPEN SOURCE
ProSponsor the shelfā™§ 0
← Back to top all-time
spdustin
ChatGPT-AutoExpert
JavaScript2023–2024
spdustin / ChatGPT-AutoExpert

ChatGPT-AutoExpert

šŸš€šŸ§ šŸ’¬ Supercharged Custom Instructions for ChatGPT (non-coding) and ChatGPT Advanced Data Analysis (coding).

18Distinct people
29Mentions
1Explicit recommenders
6.6KGitHub stars
REPOSITORY AT A GLANCE

What to know before exploring

What it is

šŸš€šŸ§ šŸ’¬ Supercharged Custom Instructions for ChatGPT (non-coding) and ChatGPT Advanced Data Analysis (coding).

GitHub

Where it is used

No documented product use was verified in the last source check.

What people are saying

AutoExpert[1] is indispensable to my use of ChatGPT now, and I’ve found that the GPT release is ā€˜better’ now than the custom instructions based release. If you haven’t used it…

kdtsh used a recommendation phrase. One indexed comment matches. This is a phrase match, not a verdict.

No classified warning is attached to this page. Read the discussions before treating silence as approval.

Also in these discussions

No other repository is repeatedly linked in the same discussions.

This week on Hacker News

Nov 9–Nov 16 UTC Ā· Direct repository links and separately labeled related project stories. Indexed through 11/16/2026, 4:59:52 AM UTC.

No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.

Mentions over time

HN mentions by month29 dated mentions

Observed mentions: 2023-10-01 18:50:49 UTC to 2024-05-26 15:49:05 UTC.

0510152023-10-01 00:00:00 UTC to 2023-11-01 00:00:00 UTC (exclusive): 7 mentions; 4 submissions, 3 commentsOct 20232023-11-01 00:00:00 UTC to 2023-12-01 00:00:00 UTC (exclusive): 15 mentions; 4 submissions, 11 comments2023-12-01 00:00:00 UTC to 2024-01-01 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-01-01 00:00:00 UTC to 2024-02-01 00:00:00 UTC (exclusive): 1 mentions; 0 submissions, 1 commentsJan 20242024-02-01 00:00:00 UTC to 2024-03-01 00:00:00 UTC (exclusive): 4 mentions; 0 submissions, 4 comments2024-03-01 00:00:00 UTC to 2024-04-01 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-04-01 00:00:00 UTC to 2024-05-01 00:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2024-05-01 00:00:00 UTC to 2024-06-01 00:00:00 UTC (exclusive): 2 mentions; 0 submissions, 2 commentsMay 2024

Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.

View mention counts and date ranges
UTC ranges include their start and exclude their end. Select a period to filter the source evidence.
PeriodStart (UTC)End (exclusive, UTC)MentionsSubmissionsComments
Oct 20232023-10-01 00:00:00 UTC2023-11-01 00:00:00 UTC743
Nov 20232023-11-01 00:00:00 UTC2023-12-01 00:00:00 UTC15411
Dec 20232023-12-01 00:00:00 UTC2024-01-01 00:00:00 UTC000
Jan 20242024-01-01 00:00:00 UTC2024-02-01 00:00:00 UTC101
Feb 20242024-02-01 00:00:00 UTC2024-03-01 00:00:00 UTC404
Mar 20242024-03-01 00:00:00 UTC2024-04-01 00:00:00 UTC000
Apr 20242024-04-01 00:00:00 UTC2024-05-01 00:00:00 UTC000
May 20242024-05-01 00:00:00 UTC2024-06-01 00:00:00 UTC202

What Hacker News says

7 sources
The "All Tools" omni-prompt takes a whopping 2,756 tokens, but it's also using the GPT-4 32k model, with a 32,767 token context window. It also reduces default DALL•E generations to one 'caption' (with a conflicting instruction later to generate 2 images). Custom Instructions are going to need some work, since all this stuff comes before CIs. The other thing that I found interesting: the built-in RAG accepts a limited number of file types, and uses the same `tools` middleware that the `browser` tool uses. It seems that it renders those files directly in their headless browser container, and captures text from it exactly the way it does when using a the Browse with Bing. It also stores those files in the `/mnt/data` sandbox used by the `python` tool, making the uploads automatically ephemeral. File types supported: "text/plain", "a
Read the rest
pplication/pdf", "text/html", "text/x-tex", "application/vnd.openxmlformats-officedocument.presentationml.presentation", "application/json", "application/vnd.openxmlformats-officedocument.wordprocessingml.document", "text/markdown" There are three middlewares (for lack of a better term) used by ChatGPT to talk to external tools. - `tools` (browser) - `tools2` (python/jupyter) - `tools3` (plugins) It's weird that DALL•E clearly uses plugins, but the model metadata endpoint [0] doesn't declare that. [0]: https://github.com/spdustin/ChatGPT-AutoExpert/blob/main/_sy...
In ChatGPT "All Tools" System omni-prompt
I’ll re-up here that I’ve collected all the ChatGPT system prompts together [0], including DALL•E 3 [1]. I wanted them all in one place as I work on the next release of my ChatGPT AutoExpert [2] ā€œcustom instructionsā€, which will include custom instructions for Voice Conversations, Vision, and yes, DALL•E. [0]: https://github.com/spdustin/ChatGPT-AutoExpert/blob/main/Sys... [1]: https://github.com/spdustin/ChatGPT-AutoExpert/blob/main/_sy... [2]: https://github.com/spdustin/ChatGPT-AutoExpert/
In DALLĀ·E 3 is now available in ChatGPT Plus and Enterprise
spdustinĀ·Oct 1, 2023Submission

Show HN: ChatGPT AutoExpert (Supercharged Custom Instructions for ChatGPT)

I’ve created two sets of ā€ChatGPT Custom Instructionsā€ for folks using ChatGPT’s web interface (i.e. not via API). They aim to dramatically improve the quality, depth, nuance, and utility of all ChatGPT responses. Both of them instruct ChatGPT to begin its completion with a preamble designed to take advantage of how the attention mechanism attends to recent tokens. They relieve the user of having to determine a task-specific prompt, and turn even the simplest questions into high quality context for ChatGPT to attend to. The first (ā€œStandard Editionā€) is for both free and paid ChatGPT subscribers. While it naturally works much better for GPT-4, it does show marked improvement for reasoning and planning, even with GPT-3.5. It automatically determines the best role to adopt, incorporates Google search links (GPT-4 is hallucination free; GPT-
Read the rest
3.5 is, well, better), and ends each response with an ā€œepilogueā€ with links to related and tangential content that the user might enjoy digging in to. The ā€œDeveloper Editionā€ is perfect for anyone using ChatGPT’s pro subscription (GPT-4 + Advanced Data Analysis), and comes with a companion Python script that you upload at the beginning of each chat. It adds /slash commands to help stash/recall text (perfect for reminding ChatGPT about your stack), building ctags (which you can ask ChatGPT to reference when writing or editing more complex code), and backing up all your files and session memory to a zip file that you can upload to a new chat to pick up where you left off. You can even upload wheels to extend the sandbox capabilities. Like the ā€œStandard Editionā€, the ā€œDeveloper Editionā€ instructs ChatGPT to begin its completion with a preamble that takes advantage of positional and semantic attention heads, influencing the quality of its generated code. The epilogue includes a brief history, as well as a simplified source/symbol tree, making it much less likely to forget what it’s been doing. It even does a great job of creating complete code, by automatically adding // TODO comments for itself, and generating code over multiple chat turns. Check them out, and tell me what you think. EDIT: I think some folks here might be particularly interested to check out how /slash commands were implemented. The technique used in the companion script is a great way to extend ChatGPT’s normally limited space for a ā€œsystem promptā€, and can easily be adopted to do prompt chaining. I’m wrapping up another script that acts like a sort of ā€œsetup wizardā€, guiding first-time users through creating a YAML file describing their stack, preferences, etc. through a series of chained prompts. I know there’s a big push to build apps that are just frontends to the API (I’ve made plenty, to be sure) but ChatGPT with Advanced Data Analysis is pretty powerful.
7 points Ā· Open discussion

MAKE YOURSELF AT HOME

A few shortcuts

/
Search the collection
?
Open this guide
Esc
Close dialogs and navigation

Your saved repositories and interests stay in this browser.