https://github.com/lmcrean/github-detective
An extension for Claude Code for identifying high-impact Github Issues
https://github.com/lmcrean/github-detective
claude-code github-api prompt-engineering
Last synced: 5 months ago
JSON representation
An extension for Claude Code for identifying high-impact Github Issues
- Host: GitHub
- URL: https://github.com/lmcrean/github-detective
- Owner: lmcrean
- Created: 2025-07-31T15:17:56.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2025-08-28T20:19:02.000Z (5 months ago)
- Last Synced: 2025-08-28T23:57:13.519Z (5 months ago)
- Topics: claude-code, github-api, prompt-engineering
- Language: Python
- Homepage:
- Size: 1.61 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
```
██████╗ ██╗████████╗██╗ ██╗██╗ ██╗██████╗
██╔════╝ ██║╚══██╔══╝██║ ██║██║ ██║██╔══██╗
██║ ███╗██║ ██║ ███████║██║ ██║██████╔╝
██║ ██║██║ ██║ ██╔══██║██║ ██║██╔══██╗
╚██████╔╝██║ ██║ ██║ ██║╚██████╔╝██████╔╝
╚═════╝ ╚═╝ ╚═╝ ╚═╝ ╚═╝ ╚═════╝ ╚═════╝
██████╗ ███████╗████████╗███████╗ ██████╗████████╗██╗██╗ ██╗███████╗
██╔══██╗██╔════╝╚══██╔══╝██╔════╝██╔════╝╚══██╔══╝██║██║ ██║██╔════╝
██║ ██║█████╗ ██║ █████╗ ██║ ██║ ██║██║ ██║█████╗
██║ ██║██╔══╝ ██║ ██╔══╝ ██║ ██║ ██║╚██╗ ██╔╝██╔══╝
██████╔╝███████╗ ██║ ███████╗╚██████╗ ██║ ██║ ╚████╔╝ ███████╗
╚═════╝ ╚══════╝ ╚═╝ ╚══════╝ ╚═════╝ ╚═╝ ╚═╝ ╚═══╝ ╚══════╝
```
# Github Detective: Bounty Hunter and Research Partner for Open Issues on Github
Locates high-impact issues on github using a combination of Github API; Python scripts and Dialogue with Claude Code CLI.
## Workflow
```
.notes/ # 1. Design Prompts with Claude Code in .notes/
scripts/ # 2. Run the python scripts in scripts
data/ # 3. data will output here
.env # recommended: use API_GITHUB_TOKEN for more tool calls
```
## How to use
Fork this repo
empty .notes dir and data (they are examples of how to use)
Interact with Claude Code as your research partner to find issues
## process overview
1. User chooses Organisation(s) e.g. Shopify, Stripe, Playwright
2. extract repositories, sort by PR velocity highest first with `30d_PR_merge_count`. Higher counts are a green flag.
3. select % of PR Velocity and filter out any repos (e.g. I don't want to work with Ruby on Rails)
4. extract open issues list, sort by recently opened
5. Define engagement criteria
6. Manually crawl through 50 most recent issues against engagement criteria (User deletes irrelevant files)
## input
input can be a CSV with
```
github org / github hyperlink / revenue (USD) / 5 tags / revenue hyperlink (confirm not a hallucination)
```
## methodology
100 Highest 2024 Revenue Orgs on github --> Repo's with Highest 30d Merge Rate, sorted by highest.money and time
the scope ends at repos to avoid complexity overheads of comparing tickets generally the repos will be manually chosen and examined from there
## output
Table output can be CSV with
```
org; repo; hyperlink; 30d_merge_rate
```