I was a bit skeptical about AI coding assistant at first. But since Skills came out, I can really feel the power of agentic workflow.
I use Cursor but these features should also work in general agents e.g. Claude Code.
Here are the features I've used
1. Plan Mode #
Plan Mode is extremely useful to explore the complexities between different solutions before making any changes.
How I used that:
- Plan
- Structure the problem first
- Avoid touching code
- Draw diagram to visualize the flow and architecture
- Execute first few steps for quick validation: Execute the initial steps to validate the approach before committing to the full implementation.
- implementation: Update the plan accordingly every round based on the real AI + human implementation.
- Futher usage: keep the plan for futher use e.g. decision making reference, QA testing instructions, tech sharing materials, work log, etc.
TODO #
- Use Planning with Files to consolidate template.
2. Agent Skills #
open standard for extending AI agents with specialized capabilities.
I used #
- vercel-react-best-practices
- perform code review
3. Rules #
Project based rules to provide contex automatically in chat.
Use cases: general coding guidelines, code structure, testing style
examples: awesome-cursorrules
4. Normal Chat / Others #
- code trace
- debugging
- writing tests
- commit message generation
Other tools to try #
- Code review in CI: Code Review with Cursor CLI
- Hooks: https://cursor.com/docs/agent/hooks
- Analyze Chat history for personal Self-Insight.