w-council — Claude Code skill: what it does, how to install
The w-council Claude Code skill explores a codebase with parallel agents. It gathers architectural findings and hands them off for follow-up work.
MIT 647 installs ★ 1 updated 2026-10-02
npx skills add blockmatic-icebox/basilic-skills --skill w-council -a claude-codeWhat w-council does
w-council first inspects the requested area, including its packages, README files, scripts, and current behavior. It then starts a small group of explorers when the investigation calls for them. Their reports are checked against files and line-level evidence; findings are verified again if the working tree changes.
The skill is for investigation, not project edits. It does not commit, push, or open a PR. After reconciling the reports, it stops or produces a plan; implementation proceeds only when the user has already requested it. Automatic invocation is disabled, so the skill does not activate on its own.
What it helps with
- map the architecture of a project area
- locate relevant packages, README files, and scripts
- check agent findings against source files
- turn investigation results into a plan
Example prompts for the agent
- Use w-council to investigate the authentication module and cite the relevant files and lines.
- Explore how payment processing works with parallel agents, then verify their findings against the code.
- Investigate this part of the codebase and draft a plan from the findings.
How to install w-council
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-council -a claude-code2Manually: put the files in place
~/.claude/skills/w-council/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/w-council/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-council -a codex2Manually: put the files in place
.agents/skills/w-council/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-council -a cursor2Manually: put the files in place
.cursor/skills/w-council/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install blockmatic-icebox/basilic-skills skills/workflow/w-council2Manually: put the files in place
.github/skills/w-council/SKILL.md1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-council -a gemini-cli2Platform CLI
gemini skills install https://github.com/blockmatic-icebox/basilic-skills --path skills/workflow/w-councilManually: put the files in place
.devin/skills/w-council/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-council -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-council
How do I install w-council in Claude Code?
Run npx skills add blockmatic-icebox/basilic-skills --skill w-council -a claude-code. The skill file is at ~/.claude/skills/w-council/SKILL.md.
What does w-council do?
It investigates a selected part of a codebase, dispatches parallel agents when needed, and checks their reports against the source tree. It then stops or creates a plan.
Does w-council change project files?
No. It does not commit, push, or open a PR. Implementation starts only if the user has already asked for it.
Source text
SKILL.md · MIT
--- name: w-council description: Explore a codebase with parallel agents, then hand off findings. disable-model-invocation: true --- Gather architecture and keywords for the requested area, then spawn a small set of varied explorers. This playbook inspects. It does not commit, push, or open a PR. Implementation continues only when the user already asked for it. 1. Inspect the area yourself first: owning packages, README/scripts, and current behavior. Record keywords and the architecture sketch the agents will use. Skip spawn when one targeted read answers the question. 2. Spawn explorers only as needed. Default to 2–3 with distinct angles, including one out-of-the-box probe. Each brief is self-contained: roots, keywords, sketch, read-only, file:line evidence, no fixes, no further spawn. Split larger investigations into separate bounded runs. 3. Reconcile reports against the tree. Prefer file evidence over agent summaries. If the working tree changed underfoot, re-verify before using a finding. 4. Investigate-only: stop after reconcile. Plan: emit [plan output](../w-plan/references/plan-output.md) in this turn. Follow [build](../w-build/SKILL.md) only if they already asked to implement — do not publish.
© blockmatic-icebox · MIT · blockmatic-icebox/basilic-skills