infragate · Minitour · Feb 13, 2026 · Feb 13, 2026 · Feb 13, 2026 · Feb 13, 2026
diff --git a/README.md b/README.md
@@ -2,29 +2,6 @@
 
 **CAPA** is a smart tool manager for AI agents that reduces context window bloat and gives agents control over their own toolset.
 
-## Why CAPA?
-
-**The Problem:** Traditional MCP setups expose all tools at once, overwhelming the agent's context window and making it harder to focus on the task at hand.
-
-**The Solution:** CAPA lets agents dynamically load only the tools they need, when they need them. One MCP server, different tools per session.
-
-### Key Benefits
-
-🎯 **Reduced Context Window**  
-Only expose relevant tools per session. The agent requests capabilities on-demand instead of having dozens of tools loaded upfront.
-
-🔄 **Multiple Profiles**  
-Define different capability profiles (e.g., "web-research", "data-analysis"). The agent switches between them automatically based on the task.
-
-✅ **No Missing Tools**  
-Define capabilities once, run everywhere. Share configurations with your team—no more "tool not found" errors across different machines.
-
-🧠 **Self-Improvement**  
-Agents can detect missing capabilities and update the capabilities file, learning which tools they need over time.
-
-💚 **Health Monitoring**  
-CAPA automatically monitors and restarts MCP servers if they crash, ensuring tools are always available.
-
 ## Installation
 
 **macOS and Linux:**
@@ -37,11 +14,9 @@ curl -LsSf https://capa.infragate.ai/install.sh | sh
 powershell -ExecutionPolicy ByPass -c "irm https://capa.infragate.ai/install.ps1 | iex"
 ```
 
-> For advanced options, see [INSTALL.md](INSTALL.md)
-
 ## Quick Start
 
-### 1. Initialize
+### Initialize your project
 
 ```bash
 cd your-project
@@ -50,12 +25,10 @@ capa init
 
 This creates a `capabilities.yaml` file where you define your agent's tools and skills.
 
-### 2. Define Capabilities
-
-Here's a minimal example with web search:
+### Example capabilities file
 
 ```yaml
-clients:
+providers:
   - cursor
 
 skills:
@@ -70,108 +43,6 @@ skills:
         Use the brave_search tool to find current information online.
     requires:
       - brave_search
-servers:
-  - id: brave
-    type: mcp
-    def:
-      cmd: npx -y @modelcontextprotocol/server-brave-search
-      env:
-        BRAVE_API_KEY: ${BraveApiKey}
-
-tools:
-  - id: brave_search
-    type: mcp
-    def:
-      server: "@brave"
-      tool: brave_web_search
-```
-
-### 3. Install
-
-```bash
-capa install
-```
-
-CAPA will:
-- Install skills to your client (e.g., `.cursor/skills/`)
-- Start the CAPA server
-- Prompt for credentials if needed (opens a web UI)
-- Display your MCP endpoint URL
-
-### 4. Configure Your MCP Client
-
-**Good news!** `capa install` automatically registers with supported clients (Cursor, Claude Desktop). 
-
-If you need to verify or manually configure, the entry in `.cursor/mcp.json` looks like:
-
-```json
-{
-  "mcpServers": {
-    "capa-your-project-id": {
-      "url": "http://localhost:5912/your-project-id/mcp"
-    }
-  }
-}
-```
-
-Note the key format is `capa-${projectId}`, allowing multiple CAPA projects in the same client.
-
-### 5. Use It
-
-The agent now sees a single tool: `setup_tools(skills: string[])`. When it calls:
-
-```typescript
-setup_tools(["web-researcher"])
-```
-
-CAPA dynamically loads the `brave_search` tool. The agent's context stays clean until tools are actually needed.
-
-## How It Works
-
-CAPA acts as an intelligent proxy between your MCP client and multiple tool sources:
-
-```
-┌─────────────┐
-│ MCP Client  │  (e.g., Cursor)
-└──────┬──────┘
-       │ Sees only: setup_tools()
-       ↓
-┌─────────────┐
-│ CAPA Server │  Manages tool lifecycle
-└──────┬──────┘
-       │ Exposes tools on demand
-       ├──→ MCP Server 1
-       ├──→ MCP Server 2
-       └──→ Command Tools
-```
-
-**Initial state:** Agent sees one tool: `setup_tools(skills: string[])`
-
-**After calling `setup_tools(["web-researcher"])`:** Agent now sees all tools defined in that skill (e.g., `brave_search`)
-
-**Benefits:**
-- Context stays clean until tools are needed
-- Agent chooses which capabilities to activate
-- Multiple skill profiles can coexist
-
-## Examples
-
-### Example 1: Web Research
-
-```yaml
-clients:
-  - cursor
-
-skills:
-  - id: web-research
-    type: inline
-    def:
-      content: |
-        ---
-        name: web-research
-        description: Search and scrape web content
-        ---
-        Use brave_search to find information and web_scrape to extract content.
 
 servers:
   - id: brave
@@ -187,161 +58,21 @@ tools:
     def:
       server: "@brave"
       tool: brave_web_search
-
-  - id: web_scrape
-    type: command
-    def:
-      init:
-        cmd: pip install beautifulsoup4 requests
-      run:
-        args:
-          - name: url
-            type: string
-            required: true
-        cmd: python -c "import requests; from bs4 import BeautifulSoup; print(BeautifulSoup(requests.get('{url}').text, 'html.parser').get_text())"
 ```
 
-### Example 2: Multiple Profiles
-
-```yaml
-clients:
-  - cursor
-
-skills:
-  - id: researcher
-    type: inline
-    def:
-      content: |
-        ---
-        name: researcher
-        ---
-        For research tasks: use brave_search
-
-  - id: data-analyst
-    type: inline
-    def:
-      content: |
-        ---
-        name: data-analyst
-        ---
-        For data analysis: use pandas_query and plot_data
-
-servers:
-  - id: brave
-    type: mcp
-    def:
-      cmd: npx -y @modelcontextprotocol/server-brave-search
-      env:
-        BRAVE_API_KEY: ${BraveApiKey}
-
-tools:
-  - id: brave_search
-    type: mcp
-    def:
-      server: "@brave"
-      tool: brave_web_search
-
-  - id: pandas_query
-    type: command
-    def:
-      init:
-        cmd: pip install pandas
-      run:
-        args:
-          - name: file
-            type: string
-          - name: query
-            type: string
-        cmd: python -c "import pandas as pd; df = pd.read_csv('{file}'); print(df.query('{query}'))"
-
-  - id: plot_data
-    type: command
-    def:
-      init:
-        cmd: pip install matplotlib pandas
-      run:
-        args:
-          - name: file
-            type: string
-          - name: x
-            type: string
-          - name: y
-            type: string
-        cmd: python -c "import pandas as pd; import matplotlib.pyplot as plt; df = pd.read_csv('{file}'); df.plot(x='{x}', y='{y}'); plt.savefig('plot.png')"
-```
-
-The agent can call `setup_tools(["researcher"])` or `setup_tools(["data-analyst"])` based on the task.
-
-## Adding Skills from GitHub
-
-You can quickly add skills from the [skills.sh](https://skills.sh) ecosystem:
-
-```bash
-# Add a skill from GitHub
-capa add vercel-labs/agent-skills
-
-# Add and customize the ID
-capa add vercel-labs/agent-skills --id my-skill
-```
-
-## CLI Reference
-
-```bash
-capa init                    # Create capabilities.yaml
-capa install                 # Install skills and start server
-capa install -e              # Install with variables from .env file
-capa install -e .prod.env    # Install with variables from custom env file
-capa add <source>            # Add skill from GitHub/GitLab/URL
-capa clean                   # Remove installed skills
-capa start                   # Start CAPA server
-capa stop                    # Stop CAPA server
-capa restart                 # Restart CAPA server
-capa status                  # Check server health status
-```
-
-## Secure Credentials
-
-Use `${VariableName}` in your config for API keys:
-
-```yaml
-servers:
-  - id: brave
-    type: mcp
-    def:
-      cmd: npx -y @modelcontextprotocol/server-brave-search
-      env:
-        BRAVE_API_KEY: ${BraveApiKey}
-```
-
-### Option 1: Web UI (Default)
-CAPA will prompt you with a web UI to enter these values securely. They're stored in a local database, never in the config file.
-
-### Option 2: Environment File
-You can also provide credentials via a `.env` file during installation:
-
-```bash
-# Create a .env file
-echo "BraveApiKey=your-api-key-here" > .env
-
-# Install with environment variables
-capa install -e
-```
-
-Or use a custom env file name:
+### Install
 
 ```bash
-capa install -e .prod.env
+capa install
 ```
 
-**Note:** If you use the `-e` or `--env` flag, the specified env file must exist, or installation will fail with a clear error message.
-
-## Learn More
+CAPA will install skills, start the server, and automatically register with your MCP client (Cursor, Claude Desktop).
 
-- **[Full Documentation](https://capa.infragate.ai)** - Complete guides and API reference
+## Documentation
 
-## Contributing
+For complete guides, examples, and API reference, visit:
 
-Contributions welcome! Open an issue or PR on [GitHub](https://github.com/infragate/capa).
+**[https://capa.infragate.ai](https://capa.infragate.ai)**
 
 ## License
 

diff --git a/icons/github.png b/icons/github.png
diff --git a/icons/gitlab.png b/icons/gitlab.png
diff --git a/src/cli/commands/auth.ts b/src/cli/commands/auth.ts
@@ -87,6 +87,45 @@ export async function authCommand(provider?: string): Promise<void> {
     process.exit(1);
   }
 
+  // Check if already authenticated with a valid token
+  const existingToken = db.getGitOAuthToken(provider);
+  if (existingToken) {
+    // Check if token is expired
+    let isExpired = false;
+    if (existingToken.expires_at) {
+      const expiresAt = new Date(existingToken.expires_at);
+      const now = new Date();
+      isExpired = expiresAt.getTime() <= now.getTime();
+    }
+
+    if (!isExpired) {
+      console.log(`✓ Already authenticated with ${provider}`);
+
+      if (existingToken.expires_at) {
+        const expiresAt = new Date(existingToken.expires_at);
+        const now = new Date();
+        const hoursUntilExpiry = Math.floor((expiresAt.getTime() - now.getTime()) / (1000 * 60 * 60));
+
+        console.log(`📅 Token expires: ${expiresAt.toLocaleString()}`);
+        console.log(`   (in approximately ${hoursUntilExpiry} hours)`);
+
+        if (existingToken.refresh_token) {
+          console.log(`✓ Refresh token available - will auto-refresh before expiration`);
+        }
+      } else {
+        console.log(`✓ Token has no expiration`);
+      }
+
+      console.log('\nTo re-authenticate, first disconnect from the provider:');
+      console.log(`  Visit: ${serverStatus.url}/ui/integrations`);
+
+      db.close();
+      return;
+    } else {
+      console.log(`⚠️  Existing token is expired, re-authenticating...\n`);
+    }
+  }
+
   console.log(`Authenticating with: ${provider}\n`);
 
   try {