w-plan — Claude Code skill: planning and installation
The w-plan Claude Code skill turns a project change into an ordered work plan. It inspects relevant context and task dependencies without modifying implementation.
MIT 647 installs ★ 1 updated 2026-10-02
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a claude-codeWhat w-plan does
w-plan explores the affected part of a project, then lays out the work needed for a requested change. It checks applicable rules and skills, inspects affected packages, README files, and scripts, and records goals, constraints, and assumptions.
Tasks are ordered and include likely files and dependencies; uncertain work comes first. The skill does not implement changes, create a branch, or commit. It writes the plan to a requested path, the repository’s existing plan file, or the chat.
For a clear, single change, the workflow points to /w-build; a blocking product gap goes to /w-clarify. Plans can be stress-tested with /w-grill, while work larger than one session goes to /w-wayfinder.
What it helps with
- turn a project change into ordered tasks
- record goals, constraints, and assumptions
- map tasks to likely files and dependencies
- save a plan to a project file or chat
Example prompts for the agent
- Plan the authorization change, including its goals, constraints, and affected files.
- Break the API migration into ordered stages and put uncertain work first.
- Draft a migration plan and save it to
docs/plan.md.
How to install w-plan
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a claude-code2Manually: put the files in place
~/.claude/skills/w-plan/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/w-plan/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a codex2Manually: put the files in place
.agents/skills/w-plan/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a cursor2Manually: put the files in place
.cursor/skills/w-plan/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install blockmatic-icebox/basilic-skills skills/workflow/w-plan2Manually: put the files in place
.github/skills/w-plan/SKILL.md1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a gemini-cli2Platform CLI
gemini skills install https://github.com/blockmatic-icebox/basilic-skills --path skills/workflow/w-planManually: put the files in place
.devin/skills/w-plan/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about w-plan
How do I install w-plan in Claude Code?
Run npx skills add blockmatic-icebox/basilic-skills --skill w-plan -a claude-code. The skill file is at ~/.claude/skills/w-plan/SKILL.md.
What does w-plan do?
It examines the project and produces an ordered plan with goals, constraints, assumptions, likely files, and task dependencies. It does not change the implementation.
Does w-plan change code or create a branch?
No. It plans the work without modifying implementation, creating a branch, or committing.
Source text
SKILL.md · MIT
--- name: w-plan description: Write an ordered plan without changing implementation. disable-model-invocation: true --- Planning does not implement, branch, or commit. A single already-clear change uses `/w-build`. A blocking product gap uses `/w-clarify`. Stress-test a plan with `/w-grill`. Fog bigger than one session uses `/w-wayfinder`. 1. **Explore**: if this turn has no reconciled exploration for the area, follow [council](../w-council/SKILL.md) steps 1–3, then continue. 2. **Goals**: 3–7 bullets for scope, success, constraints, non-goals. 3. **Context**: matching glob rule and skill, then topic MDX (or this repo’s docs path). Rules override skills. Inspect affected packages, README, scripts. Durable architecture: matching MDX/ADR. Generated outputs name an owning source; commands come from inspected scripts. 4. **Assumptions**: 3–5 bullets. Defer product scope, secrets, and destructive ops. 5. **Tasks**: ordered work with likely files and dependencies. Uncertain work first. Diagram only if relationships need one. 6. **Output**: user’s path, else the repo’s existing plan file, else chat. Follow [plan output](references/plan-output.md).
© blockmatic-icebox · MIT · blockmatic-icebox/basilic-skills