WHEN: Context setup, token optimization, MCP/agent configuration requests WHAT: Project analysis + Preset recommendation + MCP server selection + Agent activation + Config generation WHEN NOT: Status check only → /context-status command
Analyzes your project and recommends optimized context presets to save tokens. Activates when you request context setup or start a new project, then generates MCP server and agent configurations.
/plugin marketplace add physics91/claude-vibe/plugin install claude-vibe@physics91-pluginsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Optimizes Claude Code's context window by interactively selecting MCP servers, Custom Agents, and Slash Commands per project and generating configuration files.
Activate when:
/context-setup command is executedAnalyze project root files to detect project type.
Analysis targets:
package.json (Node.js/JavaScript)requirements.txt, pyproject.toml (Python)go.mod (Go)Cargo.toml (Rust)pom.xml, build.gradle (Java)Recommend preset based on detected project type.
AskUserQuestion:
"Select a context preset for your project"
Options:
- Web Development - React/Vue/Next.js (~28k saved)
- Python Web - FastAPI/Django/Flask (~30k saved)
- Go Backend - Gin/Echo/Fiber (~28k saved)
- Rust Systems - Actix/Axum (~32k saved)
- DevOps - Docker/K8s/Terraform (~25k saved)
- Data Science - ML/AI/Data (~30k saved)
- Full Stack - Web + API + DB (~20k saved)
- Minimal - Core tools only (~45k saved)
- Custom - Select manually
If Custom is selected, choose by category.
MCP Server Selection:
AskUserQuestion: "Select MCP servers to enable"
Options:
- GitHub (PR/issue management) - ~8k tokens
- Playwright (browser automation) - ~12k tokens
- Brave Search (web search) - ~3k tokens
- Select all
multiSelect: true
Agent Category Selection:
AskUserQuestion: "Which agent categories do you need?"
Options:
- Frontend (React, Vue, Next.js, CSS)
- Python (FastAPI, Django, Flask, Data)
- Go/Rust (Go API, Rust Systems)
- Backend (API, DB, Security)
- DevOps (Docker, K8s, Terraform, CI/CD)
- Database (SQL, ORM, Migrations)
- Mobile (iOS, Android, Flutter)
- AI/ML (ML, DL, NLP)
multiSelect: true
Create based on selections:
.claude/context-profile.json: Project context settings.claude/.mcp.json: MCP server settings (selected servers only)MCP settings apply after session restart.
Setup complete!
Generated files:
- .claude/context-profile.json
- .claude/.mcp.json
Active MCP servers: github, playwright, brave-search
Active agents: react-expert, css-expert, nodejs-expert
Estimated token savings: ~28,000 tokens (14%)
Restart Claude Code to apply MCP settings.
Project analyzed.
**Detected project type**: Web Development (React + TypeScript)
**Confidence**: 85%
**Recommended preset**: web-dev
**Estimated token savings**: ~28,000 tokens
[Present preset selection with AskUserQuestion]
Context setup complete!
**Profile**: Web Development
**Project**: G:\ai-dev\my-project
### Active MCP Servers (3/6)
- github
- playwright
- brave-search
### Active Agents (5/55)
- react-expert
- css-expert
- nodejs-expert
- frontend-optimizer
- ui-ux-designer
### Generated Files
- .claude/context-profile.json
- .claude/.mcp.json
### Estimated Token Savings
~28,000 tokens (14% saved)
Restart Claude Code to apply MCP settings.
To change settings, run `/context-setup`.
## Current Context Status
**Profile**: Web Development
**Project**: G:\ai-dev\my-project
### MCP Servers (3/6 active)
[x] github
[x] playwright
[x] brave-search
[ ] filesystem (inactive)
[ ] openrouter (inactive)
[ ] context7 (inactive)
### Agents (5/55 active)
[x] react-expert, css-expert, nodejs-expert, frontend-optimizer, ui-ux-designer
### Estimated Token Savings
~28,000 tokens
This skill integrates with:
SessionStart hook: Check project profile existence and notify/context-setup command: Activate this skill/context-status command: Display current status| Preset | Description | MCP Servers | Token Savings |
|---|---|---|---|
| minimal | Core tools only | None | ~45k |
| web-dev | React/Vue/Next.js | github, playwright, brave-search | ~28k |
| python-web | FastAPI/Django/Flask | github, brave-search, context7 | ~30k |
| go-backend | Gin/Echo/Fiber/Chi | github, brave-search, context7 | ~28k |
| rust-systems | Actix-web/Axum | github, brave-search, context7 | ~32k |
| devops | Docker/K8s/Terraform | github, brave-search, context7 | ~25k |
| api-dev | Backend API | github, brave-search, context7 | ~25k |
| data-science | ML/AI/Data | brave-search, context7, github | ~30k |
| full-stack | Web + API + DB | github, playwright, brave-search, context7 | ~20k |
.claude/.mcp.json takes precedence over global settingsThis skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.