- Home
- Skills
- Proxiblue
- Claude Skills
- Github Analysis
github-analysis_skill
7
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill proxiblue/claude-skills --skill github-analysis- SKILL.md12.6 KB
Overview
This skill systematically analyzes GitHub tickets by reproducing reported issues in a local DDEV environment before proposing fixes. It automates production-to-local URL substitution, ensures database parity, and fetches missing media so investigations are accurate. The goal is to avoid wasted debugging on cache, data, or environment discrepancies.
How this skill works
The skill reads the ticket to extract URLs, reproduction steps, screenshots, and expected vs actual behavior, then converts production URLs to the local DDEV host. It verifies DDEV is running, compares local and production pages, and—when needed—pulls a filtered production database and syncs media assets. Once the issue reproduces locally, it guides root cause analysis, tests fixes, and documents the solution with clear testing and deployment notes.
When to use it
- User asks to analyze, debug, fix, or investigate a GitHub issue
- User provides an issue number or production URL to investigate
- Reproduction requests that mention missing products, orders, or media
- When local behavior differs from production and data-sync may be required
- When screenshots show missing images or 404 media on local site
Best practices
- Never propose a code fix before reproducing the issue locally
- Always convert production URLs to the DDEV host format (https://{project}.ddev.site)
- Pull a filtered production database if local content differs or specific records are missing
- Download or rsync missing media assets when images or files return 404 locally
- Clear caches and reindex after importing data, then retest the exact reproduction steps
Example use cases
- Ticket: product page price mismatch — convert URL, pull DB, reproduce and confirm stale local data
- Ticket: missing product images — fetch specific images or rsync pub/media and verify paths
- Ticket: customer order workflow failing — import filtered DB, reproduce using same customer, inspect logs
- Ticket: JavaScript console errors only on production — reproduce locally and enable Xdebug to inspect back-end errors
FAQ
Document the limitations, request secure access or an exported filtered snapshot from a production operator, and attempt reproduction with available data while noting gaps.
When should I sync the entire media directory vs individual files?
Sync individual files when a few assets are missing; use rsync for large-scale missing media or when many pages show broken images, but plan for transfer size and permissions.