# NB-Corp/gpt-with-taste

> A set of system instructions and hooks for Codex that steer GPT's working style: sensible defaults, careful delivery, and engineering/visual craft. Installs into CODEX_HOME without changing model, provider, or permissions.

- Magnitude: 4.6 out of 10 — Early signal
- Stars: 112 total · +10 stars measured 2026-10-15, 00:21–02:29 UTC, ≈ 81 by evening
- Star trust: star growth looks organic
- Category: AI agents · Language: JavaScript · Created: 2026-10-12 · Last push: 2026-10-13
- GitHub: https://github.com/NB-Corp/gpt-with-taste · Page: https://gitnova.dev/en/r/NB-Corp/gpt-with-taste

## Useful for

- Install the recipe into Codex via node scripts/install.mjs --hooks for multi-turn work
- Steer GPT's behavior when writing docs and code without editing AGENTS.md
- Compare model outputs before and after install using the Case Archive examples

## Why it’s here

- Star-counter measurements on 2026-10-15 (UTC), 00:21–02:29: 102 → 112 stars (+10). This is the change over that interval.
- Estimated end-of-day forecast: about +81 stars, using observed gains and the previous day.
- The repository is 3 days old and already has 112 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.
- Top new repositories this week: #151.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 8
- Issues and pull requests: 0
- Watchers: 0
- Average over the last week: 50 per day
- Usual pace: too little history (under two weeks)
- Stars in the last hour (measured): 6

## Stars per day, last 4 days (oldest → newest, today is partial)

2026-10-12 … 2026-10-15: 18, 23, 79, 10

## Spotted in now

- Top new repositories this week: #151

## More in this category

1. **morluto/rea** — 7.3 · Breakout · AI agents · TypeScript · +3,436 stars measured 2026-10-15, 00:20–02:28 UTC, ≈ 27,746 by evening
   A reverse-engineering tool that lets AI agents inspect apps and native binaries: it connects agents to Hopper or Ghidra via MCP and CLI, helping explain how a feature works and rebuild it in your own project.
   Full card: https://gitnova.dev/en/r/morluto/rea.md
2. **outcrawlai/scout** — 7.2 · Early signal · AI agents · TypeScript · +16 stars measured 2026-10-15, 00:20–02:28 UTC, ≈ 553 by evening
   Open-source browser agent in TypeScript: takes a task in plain words, opens a real browser, reads pages, clicks, types and scrolls until the job is done, then returns the answer. Works with any chat completions API service or a local model.
   Full card: https://gitnova.dev/en/r/outcrawlai/scout.md
3. **i12bp8/decomp.wtf** — 6.4 · Early signal · AI agents · Python · +19 stars measured 2026-10-15, 00:21–02:28 UTC, ≈ 209 by evening
   decomp.wtf points an existing AI coding agent at a game ROM you own, hands it one function at a time, compiles each attempt with the game's original compiler and checks the result byte for byte. It runs locally and supports N64, PS1/PS2,…
   Full card: https://gitnova.dev/en/r/i12bp8/decomp.wtf.md
4. **Graphicle-AI/graphicle** — 6.4 · Early signal · AI agents · Rust · +7 stars measured 2026-10-15, 00:21–02:28 UTC, ≈ 116 by evening
   Graphicle is a Rust program that builds a code graph of your repository and stores project memory (docs, decisions, past agent sessions), returning cited excerpts with exact files and lines to coding agents. It runs locally over MCP with…
   Full card: https://gitnova.dev/en/r/Graphicle-AI/graphicle.md
5. **onehumanbeing/esp32-iphone-agent** — 6.1 · Peaking · AI agents · C · +37 stars measured 2026-10-15, 00:20–02:28 UTC, ≈ 346 by evening
   Lets an AI model drive a real iPhone via an ESP32-S3: the screen is mirrored over AirPlay while the board emulates a USB mouse and keyboard. No Mac, Xcode, Apple developer account or jailbreak required.
   Full card: https://gitnova.dev/en/r/onehumanbeing/esp32-iphone-agent.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-15 02:41 UTC, updated every 30 minutes.