Plan a feature before Claude writes code
For anyone building with Claude Code who has watched it charge off in the wrong direction: get a one-page plan first, approve it, then build in small checked steps. Claude reads the relevant code, writes the goal, what done looks like, the files to touch, numbered steps each with a check, and the risks and questions, then stops for your yes. It builds one step at a time, runs typecheck and tests after each, ticks the plan, and tells you what changed from it.
SkillCodingv1.0.0 (latest)
- Activity
- New: be the first to try it
- Installs
- 0 installs
- Likes
- 0 likes
- Clones
- 0 clones
How this works
- It is a skill: written instructions Claude reads when your request matches what the skill is for. You do not call it by name; just ask for the task.
- While it works, it reads and changes files in the places it lists, nowhere else.
- Nothing runs until you choose to use it, and you can remove it any time.
What it can access
What this kit declares, in the same words the CLI shows before you install. On Claude Code the CLI turns these into permission rules, so the agent stops anything undeclared.
Reads
Reads files in ./**
Writes
Writes files in ./docs/plans/**
Writes
Writes files in ./out/plans/**