I will set up your ai coding assistant for your project with rules and commands


Über diesen Service
Tired of your AI assistant ignoring your code style, folder layout and your "never commit to main" rule?
I set up your AI coding assistant so it knows your project - and I do it for WHICHEVER tool you use: Claude Code, Cursor, MiniMax, DeepSeek, Codex or anything else that reads a project context file.
WHAT YOU GET
- A tailored project context file (AGENTS.md / CLAUDE.md / .cursorrules - your choice) with your stack, exact build / test / lint commands, code conventions and an explicit Do / Don't list
- The matching config folder for your tool (.claude/, .cursor/, .minimax/ etc.) in that tool's native format - not a renamed copy of someone else's
- Custom commands for your repeat tasks, written in YOUR tool's format
- A permission allowlist for routine commands, plus a deny-list that blocks .env, secrets, git push and rm -rf
- A README handoff so you can maintain it without me
WHY ME
I ship AI tooling for a living: a Chrome extension with a Cloudflare Worker licence API, a serverless AI backend on Cloudflare Workers, and my own project-context packs.
Any stack: Node/React, Python, Go, Rust, PHP and more. Message me before ordering with your tool, stack and goals.
Lerne Jiaming kennen
Custom HR Software, Employee Management, Inventory, Cloudflare and AI Automation
- AusChina
- Mitglied seitJuli 2026
Sprachen
Chinesisch, Englisch
Mein Portfolio
FAQ
Can you set it up for my specific tool?
Yes - tell me the tool and version and I deliver that tool's native format, not a renamed copy of another tool's config. For the full Claude Code feature set (hooks, subagents, MCP wiring), order my Claude Code gig.
What if my team uses more than one AI tool?
That is fine. One shared context file covers the rules and I add the small per-tool config each one needs. Just tell me the list.
Will this really stop the AI ignoring my rules?
Mostly. The context file is read automatically by tools that support it, and the commands give a repeatable entry point for your daily tasks. Not magic, but it removes the most common complaint.
Do I need to give you repo access?
Not necessarily - I can work from a description of your stack and conventions. Read-only access helps for a deeper setup, but it is your call.
How fast will you reply and start?
I reply within 3-4 hours, usually the same working day. You get my recommended approach and a clear price before any work starts. If your task needs more time or a different approach, I will say so straight away instead of going quiet.

