Skip to content

Commit 4b3a566

Browse files
Robbie1977Robbie1977
committed
Bump version to 1.4.4
- Updated version from 1.4.2 to 1.4.4 - Enhanced documentation examples across all files - Fixed Claude Desktop MCP configuration format - Improved testing examples with comprehensive workflows - Updated Gemini setup instructions
1 parent 82a6d7f commit 4b3a566

4 files changed

Lines changed: 99 additions & 60 deletions

File tree

‎README.md‎

Lines changed: 48 additions & 26 deletions
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,9 @@ The easiest way to use VFB3-MCP is through our hosted service at `https://vfb3-m
2222
{
2323
"mcpServers": {
2424
"virtual-fly-brain": {
25-
"url": "https://vfb3-mcp.virtualflybrain.org"
25+
"type": "http",
26+
"url": "https://vfb3-mcp.virtualflybrain.org",
27+
"tools": ["*"]
2628
}
2729
}
2830
}
@@ -39,7 +41,9 @@ The easiest way to use VFB3-MCP is through our hosted service at `https://vfb3-m
3941
{
4042
"mcpServers": {
4143
"virtual-fly-brain": {
42-
"url": "https://vfb3-mcp.virtualflybrain.org"
44+
"type": "http",
45+
"url": "https://vfb3-mcp.virtualflybrain.org",
46+
"tools": ["*"]
4347
}
4448
}
4549
}
@@ -56,6 +60,18 @@ The easiest way to use VFB3-MCP is through our hosted service at `https://vfb3-m
5660
5. **Add the server URL**: `https://vfb3-mcp.virtualflybrain.org`
5761
6. **Give it a name** like "Virtual Fly Brain"
5862

63+
**Alternative JSON configuration** (in `mcp.json`):
64+
```json
65+
{
66+
"servers": {
67+
"virtual-fly-brain": {
68+
"type": "http",
69+
"url": "https://vfb3-mcp.virtualflybrain.org"
70+
}
71+
}
72+
}
73+
```
74+
5975
#### Visual Studio Code (with MCP Extension)
6076

6177
1. **Install the MCP extension** for VS Code from the marketplace
@@ -75,52 +91,58 @@ For any MCP-compatible client that supports HTTP servers:
7591
{
7692
"mcpServers": {
7793
"virtual-fly-brain": {
94+
"type": "http",
7895
"url": "https://vfb3-mcp.virtualflybrain.org",
79-
"type": "http"
96+
"tools": ["*"]
8097
}
8198
}
8299
}
83100
```
84101

85102
#### Gemini Setup
86103

87-
To use the Virtual Fly Brain (VFB) Model Context Protocol (MCP) server with AI on Google Search, connect through the Gemini CLI or a custom Python/Node.js client.
104+
To use the Virtual Fly Brain (VFB) Model Context Protocol (MCP) server with Google Gemini, you can connect through custom Python/Node.js clients that support MCP.
88105

89-
The Gemini web interface does not directly support custom MCP integration. Developer tools are needed to connect the two.
106+
**Note**: Direct Gemini web interface integration with MCP is not currently supported. Developer tools are needed to connect the two.
90107

91-
**Option 1: Using Gemini CLI**
108+
**Option 1: Using Python**
92109

93-
The Gemini CLI allows direct registration of remote MCP servers.
110+
For application development, use the `mcp` and `google-genai` libraries to connect.
94111

95-
Install the CLI: Ensure Node.js is installed, then run:
96-
```bash
97-
npm install -g @google/gemini-cli
98-
```
112+
Setup: `pip install google-genai mcp`
99113

100-
Add the VFB Server: Use the add command with the VFB URL.
101-
```bash
102-
gemini mcp add vfb https://vfb3-mcp.virtualflybrain.org
103-
```
114+
Implementation: Use an `SSEClientTransport` to connect to the VFB URL, list its tools, and pass their schemas to the Gemini model as Function Declarations.
104115

105-
Verify & Use: Run the CLI by typing `gemini`. Check the connection with `/mcp`. AI on Google Search will then call VFB's neuroanatomy and connectivity tools when questions about Drosophila are asked.
116+
#### Testing the Connection
106117

107-
**Option 2: Using Python**
118+
Once configured, you can test that VFB3-MCP is working by asking your AI assistant questions like:
108119

109-
For application development, use the `mcp` and `google-genai` libraries to connect.
120+
**Basic Queries:**
121+
- "Get information about the neuron VFB_jrcv0i43"
122+
- "Search for terms related to medulla in the fly brain"
123+
- "What neurons are in the antennal lobe?"
110124

111-
Setup: `pip install google-genai mcp`
125+
**Advanced Queries:**
126+
- "Find all neurons that connect to the mushroom body"
127+
- "Show me expression patterns for gene repo"
128+
- "What brain regions are involved in olfactory processing?"
129+
- "Run a connectivity analysis for neuron VFB_00101567"
112130

113-
Implementation: Use an `SSEClientTransport` to connect to the VFB URL, list its tools, and pass their schemas to the Gemini model as Function Declarations.
131+
**Search Examples:**
132+
- "Search for adult neurons in the visual system"
133+
- "Find genes expressed in the central complex"
134+
- "Show me all templates available in VFB"
114135

115-
#### Testing the Connection
136+
If you see responses with VirtualFlyBrain data, including neuron names, brain regions, gene expressions, or connectivity information, the setup is successful!
116137

117-
Once configured, you can test that VFB3-MCP is working by asking your AI assistant to:
138+
For more detailed usage examples and API calls, see **[examples.md](examples.md)**.
118139

119-
- "Get information about the term VFB_jrcv0i43"
120-
- "Search for terms related to medulla"
121-
- "Run a PaintedDomains query for VFB_00101567"
140+
### Example Workflow
122141

123-
If you see responses with VirtualFlyBrain data, the setup is successful!
142+
1. **Search for a term**: "Search for neurons in the optic lobe"
143+
2. **Get detailed info**: "Get information about VFB_00101567"
144+
3. **Run specific queries**: "Show connectivity for VFB_00101567"
145+
4. **Explore relationships**: "What neurons synapse in the mushroom body?"
124146

125147
## 🛠️ Local Installation
126148

‎examples.md‎

Lines changed: 6 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -96,7 +96,9 @@ Add to your MCP server configuration:
9696
{
9797
"mcpServers": {
9898
"vfb3-mcp": {
99-
"url": "https://vfb3-mcp.virtualflybrain.org"
99+
"type": "http",
100+
"url": "https://vfb3-mcp.virtualflybrain.org",
101+
"tools": ["*"]
100102
}
101103
}
102104
}
@@ -108,7 +110,9 @@ Add to your `claude.json`:
108110
{
109111
"mcpServers": {
110112
"vfb3-mcp": {
111-
"url": "https://vfb3-mcp.virtualflybrain.org"
113+
"type": "http",
114+
"url": "https://vfb3-mcp.virtualflybrain.org",
115+
"tools": ["*"]
112116
}
113117
}
114118
}

‎package.json‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
{
22
"name": "vfb3-mcp",
3-
"version": "1.4.2",
3+
"version": "1.4.4",
44
"description": "MCP server for VirtualFlyBrain API integration",
55
"main": "index.js",
66
"scripts": {

‎src/index.ts‎

Lines changed: 44 additions & 31 deletions
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@ import cors from 'cors';
1313
import express from 'express';
1414
import { randomUUID } from 'node:crypto';
1515

16-
const VERSION = '1.4.2';
16+
const VERSION = '1.4.4';
1717

1818
// GA4 Analytics configuration
1919
const GA_MEASUREMENT_ID = process.env.GA_MEASUREMENT_ID || 'G-K7DDZVVXM7';
@@ -495,7 +495,9 @@ function getHtmlPage(): string {
495495
<pre><code>{
496496
"mcpServers": {
497497
"virtual-fly-brain": {
498-
"url": "https://vfb3-mcp.virtualflybrain.org"
498+
"type": "http",
499+
"url": "https://vfb3-mcp.virtualflybrain.org",
500+
"tools": ["*"]
499501
}
500502
}
501503
}</code></pre>
@@ -515,7 +517,9 @@ function getHtmlPage(): string {
515517
<pre><code>{
516518
"mcpServers": {
517519
"virtual-fly-brain": {
518-
"url": "https://vfb3-mcp.virtualflybrain.org"
520+
"type": "http",
521+
"url": "https://vfb3-mcp.virtualflybrain.org",
522+
"tools": ["*"]
519523
}
520524
}
521525
}</code></pre>
@@ -555,50 +559,59 @@ function getHtmlPage(): string {
555559
<pre><code>{
556560
"mcpServers": {
557561
"virtual-fly-brain": {
562+
"type": "http",
558563
"url": "https://vfb3-mcp.virtualflybrain.org",
559-
"type": "http"
564+
"tools": ["*"]
560565
}
561566
}
562567
}</code></pre>
563568
</div>
564569
565570
<h3>Gemini Setup</h3>
566-
<p>To use the Virtual Fly Brain (VFB) Model Context Protocol (MCP) server with AI on Google Search, connect through the Gemini CLI or a custom Python/Node.js client.</p>
567-
<p>The Gemini web interface does not directly support custom MCP integration. Developer tools are needed to connect the two.</p>
568-
569-
<h4>Option 1: Using Gemini CLI</h4>
570-
<p>The Gemini CLI allows direct registration of remote MCP servers.</p>
571-
<ol>
572-
<li class="step"><strong>Install the CLI</strong>: Ensure Node.js is installed, then run:</li>
573-
</ol>
574-
<div class="config-json">
575-
<pre><code>npm install -g @google/gemini-cli</code></pre>
576-
</div>
577-
<ol start="2">
578-
<li class="step"><strong>Add the VFB Server</strong>: Use the add command with the VFB URL.</li>
579-
</ol>
580-
<div class="config-json">
581-
<pre><code>gemini mcp add vfb https://vfb3-mcp.virtualflybrain.org</code></pre>
582-
</div>
583-
<ol start="3">
584-
<li class="step"><strong>Verify & Use</strong>: Run the CLI by typing <code>gemini</code>. Check the connection with <code>/mcp</code>. AI on Google Search will then call VFB's neuroanatomy and connectivity tools when questions about Drosophila are asked.</li>
585-
</ol>
586-
587-
<h4>Option 2: Using Python</h4>
571+
<p>To use the Virtual Fly Brain (VFB) Model Context Protocol (MCP) server with Google Gemini, you can connect through custom Python/Node.js clients that support MCP.</p>
572+
<p><strong>Note</strong>: Direct Gemini web interface integration with MCP is not currently supported. Developer tools are needed to connect the two.</p>
573+
574+
<h4>Using Python</h4>
588575
<p>For application development, use the <code>mcp</code> and <code>google-genai</code> libraries to connect.</p>
589576
<ol>
590577
<li class="step"><strong>Setup</strong>: <code>pip install google-genai mcp</code></li>
591578
<li class="step"><strong>Implementation</strong>: Use an <code>SSEClientTransport</code> to connect to the VFB URL, list its tools, and pass their schemas to the Gemini model as Function Declarations.</li>
592579
</ol>
593580
594581
<h2>🧪 Testing the Connection</h2>
595-
<p>Once configured, you can test that VFB3-MCP is working by asking your AI assistant to:</p>
582+
<p>Once configured, you can test that VFB3-MCP is working by asking your AI assistant questions like:</p>
583+
584+
<h3>Basic Queries:</h3>
585+
<ul>
586+
<li>"Get information about the neuron VFB_jrcv0i43"</li>
587+
<li>"Search for terms related to medulla in the fly brain"</li>
588+
<li>"What neurons are in the antennal lobe?"</li>
589+
</ul>
590+
591+
<h3>Advanced Queries:</h3>
596592
<ul>
597-
<li>"Get information about the term VFB_jrcv0i43"</li>
598-
<li>"Search for terms related to medulla"</li>
599-
<li>"Run a PaintedDomains query for VFB_00101567"</li>
593+
<li>"Find all neurons that connect to the mushroom body"</li>
594+
<li>"Show me expression patterns for gene repo"</li>
595+
<li>"What brain regions are involved in olfactory processing?"</li>
596+
<li>"Run a connectivity analysis for neuron VFB_00101567"</li>
600597
</ul>
601-
<p>If you see responses with VirtualFlyBrain data, the setup is successful!</p>
598+
599+
<h3>Search Examples:</h3>
600+
<ul>
601+
<li>"Search for adult neurons in the visual system"</li>
602+
<li>"Find genes expressed in the central complex"</li>
603+
<li>"Show me all templates available in VFB"</li>
604+
</ul>
605+
606+
<p>If you see responses with VirtualFlyBrain data, including neuron names, brain regions, gene expressions, or connectivity information, the setup is successful!</p>
607+
608+
<h3>Example Workflow</h3>
609+
<ol>
610+
<li><strong>Search for a term</strong>: "Search for neurons in the optic lobe"</li>
611+
<li><strong>Get detailed info</strong>: "Get information about VFB_00101567"</li>
612+
<li><strong>Run specific queries</strong>: "Show connectivity for VFB_00101567"</li>
613+
<li><strong>Explore relationships</strong>: "What neurons synapse in the mushroom body?"</li>
614+
</ol>
602615
603616
<h2>🛠️ Available Tools</h2>
604617
<ul>

0 commit comments

Comments
 (0)