agentleFS
vibe-cokitvibe-cokit/claude-code/skills/vk-bootstrap-auto/SKILL.md

vk-bootstrap-auto

Skill1 starsChanged 4 months agoApache-2.0View on GitHub

⚡⚡⚡⚡ Bootstrap a new project automatically

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    First: if Git is not initialized, init it via git-manager (branch main).

    line 19

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Bootstrap Auto

Greenfield project orchestrator. Sets up a NEW project, then delegates implementation to /vk-cook — it does NOT re-implement the build/test/review lifecycle (that lives in /vk-cook + the Orchestration Protocol & Development Rules at workflows/*).

Ultrathink the architecture. Be brutally honest about trade-offs (Role Core + Voice in development-rules.md). Operate by YAGNI/KISS/DRY.

<user-requirements>$ARGUMENTS</user-requirements>

Modifiers (flag-literalism): --fast → lighter research, fewer approval pauses. --no-worktree → forward to /vk-cook in Step 5: skip session worktree isolation + auto-PR (workflows/worktree-session.md §1-§2), build directly in the current checkout. --no-hoh → forward to /vk-cook: opt out of HoH continual-improvement loop (cook step 4 default); --loops N → HoH iteration budget. Default (absent) = worktree isolation + HoH loop ON, so bootstrap's build step also gets an isolated branch + auto-PR + continual improvement at finalize.

Workflow

First: if Git is not initialized, init it via git-manager (branch main).

1. Research

Parallel researcher agents explore the request, validate the idea, surface challenges (cited reports; length cap per the research skill).

2. Tech Stack

planner + researcher agents pick a best-fit stack; if the user named a stack, use it. Write the stack into ./docs. Get user approval (skip extra confirmation under --fast).

3. Wireframe & Design (if UI project)

ui-ux-designer + researcher agents produce design guidelines (./docs/design-guidelines.md) + HTML wireframes (./docs/wireframe/). Research style/trends/fonts/colors/spacing. Predict actual Google Font names (not just Inter/Poppins). Assets and visual verification: workflows/visual-input.md (capture wireframe screenshots with chrome-devtools, then Read them). Get user approval.

4. Plan

planner creates a progressive-disclosure plan (plans/YYYYMMDD-HHmm-name/plan.md + phase-XX-*.md), with a file-ownership matrix (schema: workflows/parallel-execution.md) where phases can run in parallel. Get user approval.

5. Build → Test → Review → Finalize

Run /vk-cook <plan-path> --auto [--no-worktree] [--no-hoh] [--loops N] (forward --no-worktree, --no-hoh, --loops if present in <user-requirements>) to implement, test (Evidence Gate), review (Findings Gate), and finalize the plan. /vk-cook delegates the HoH loop by default for greenfield incremental growth. Parallel execution is auto-detected from the plan's file-ownership matrix (schema: workflows/parallel-execution.md).

6. Documentation

docs-manager creates/updates ./docs/{README.md (≤300 lines), project-overview-pdr.md, code-standards.md, system-architecture.md}. project-manager creates ./docs/project-roadmap.md.

7. Onboarding & Final Report

Guide setup one question at a time (e.g. obtaining API keys → env vars). Concise final report; offer to commit/push via git-manager.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.