vc:kit
Skill0 starsChanged 3 months agoMITView on GitHub
Spec-driven coding-agent harness. Sub-skill index for the vibecode-pro-max kit. Use this file to discover the right sub-skill for a task; then load that sub-skill's SKILL.md for the contract.
Files
vibecode-pro-max Kit
Spec-driven coding-agent harness. The kit is a directory of sub-skills, each with its own SKILL.md. This file is the index — when you know which sub-skill you need, load that sub-skill's SKILL.md for the contract.
How to use this file: identify the category that matches your task below, load the sub-skill's SKILL.md (e.g. cat .claude/skills/vc-scout/SKILL.md), and follow its workflow.
Thinking & Analysis
| Sub-skill | Use when |
vc-scout | Need to find files, discover code structure, or scope a search across the repo |
vc-sequential-thinking | Need to break a complex problem into ordered reasoning steps |
vc-problem-solving | Stuck on a problem and need a structured unblocking technique |
vc-watzup | Need a read-only handoff summary (active plan, branch state, recent work) |
vc-xia | Need to study how another repo solves a problem before adapting it here |
vc-debug | Investigating a bug, performance issue, or failed deployment |
vc-security | Auditing for security issues (STRIDE, OWASP) before merge |
Context & Documentation
| Sub-skill | Use when |
vc-generate-context | Generating or updating process/context/all-context.md for a project |
vc-context-engineering | Reasoning about context budgets, degradation, or optimization |
vc-docs-seeker | Need current library/framework docs via context7 |
vc-setup | First-time setup of the harness in a new project |
Plan, Build, & Release
| Sub-skill | Use when |
vc-generate-plan | Authoring a *_PLAN_*.md artifact in the canonical format |
vc-update | Pulling latest harness improvements into a consuming project |
vc-publish | (Maintainer) pushing local harness changes back to the remote kit repo |
vc-merge-worktree | Merging a finished worktree branch back into main |
Audit & Validation
| Sub-skill | Use when |
vc-audit | Validating context routing (--context), harness health (--vc), or plan inventory (--plans) |
vc-web-testing | Setting up Playwright/Vitest/k6 test scaffolding or running test patterns |
vc-stress-test | Pre-implementation risk surface — --predict (5-persona debate) or --scenario (12-dimension edge cases) |
vc-autoresearch | Running an autonomous metric-optimization loop with verify/guard/scope |
Browser & UI
| Sub-skill | Use when |
vc-chrome-devtools | Browser automation via Puppeteer (navigation, ARIA snapshots, screenshots, performance, Browserbase cloud) |
vc-preview | Generating HTML recaps, slide decks, diagrams, or view-mode toggles |
vc-tech-graph | Producing a durable publish-grade SVG/PNG diagram for process/general-plans/references/ or feature-scoped equivalents |
Repo & Data
| Sub-skill | Use when |
vc-repomix | Packing a remote repo into a single context file for review or study |
vc-mcp-management | Managing MCP server connections, capability discovery, and tool selection |
Workflow Cheat-Sheet
| I need to… | Use |
| Start a new project | vc-setup |
| Write a new plan | vc-generate-plan |
| Find code or structure | vc-scout |
| Audit the kit | vc-audit |
| Stress-test a proposal | vc-stress-test --predict |
| Find edge cases | vc-stress-test --scenario |
| Debug a bug | vc-debug |
| Run a focused test suite | vc-web-testing |
| Build a diagram | vc-tech-graph |
| Automate a browser | vc-chrome-devtools |
| Pull upstream harness updates | vc-update |
| Publish local harness changes | vc-publish |
Top-Level Conventions
- Agents live in
.claude/agents/*.mdand are loaded by name (e.g.vc-execute-agent). - Skills live in
.claude/skills/*/SKILL.mdand are loaded on demand. - Plans live in
process/general-plans/active/orprocess/features/{feature}/active/. - Context lives in
process/context/all-context.md(router) and grouped docs. - The Codex mirror (
.codex/agents/*.toml) is generated from.claude/agents/*.mdbynode .claude/skills/vc-audit/scripts/build-codex-toml.mjs. Hand-edited TOML is rejected byvc-audit/scripts/validate-agent-parity.mjs --strict.
For the broader RIPER-5 protocol and the orchestrator's role, see CLAUDE.md and AGENTS.md at the repo root.
In this repository51 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- vc:auditSkill
- vc:autoresearchSkill
- vc:chrome-devtoolsSkill
- vc:context-engineeringSkill
- vc:debugSkill
- vc:docs-seekerSkill
- vc:generate-contextSkill
- vc:generate-planSkill
- vc:mcp-managementSkill
- vc:merge-worktreeSkill
- vc:previewSkill
- vc:problem-solvingSkill
- vc:publishSkill
- vc:repomixSkill
- vc:scoutSkill
- vc:securitySkill
- vc:sequential-thinkingSkill
- vc:setupSkill
- vc:stress-testSkill
- vc:tech-graphSkill
- vc:updateSkill
- vc:watzupSkill
- vc:web-testingSkill
- vc:xiaSkill
- vc:kitSkill
- vc:auditSkill
- vc:autoresearchSkill
- vc:chrome-devtoolsSkill
- vc:context-engineeringSkill
- vc:debugSkill
- vc:docs-seekerSkill
- vc:generate-contextSkill
- vc:generate-planSkill
- vc:mcp-managementSkill
- vc:merge-worktreeSkill
- vc:previewSkill
- vc:problem-solvingSkill
- vc:publishSkill
- vc:repomixSkill
- vc:scoutSkill
- vc:securitySkill
- vc:sequential-thinkingSkill
- vc:setupSkill
- vc:stress-testSkill
- vc:tech-graphSkill
- vc:updateSkill
- vc:watzupSkill
- vc:web-testingSkill
- vc:xiaSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.