Enables collaboration with Google's Gemini AI to provide technical advice, code reviews, and iterative development support, including PRD creation and tech stack decision-making.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Gemini Collaboration MCP Server Review this code and suggest potential performance improvements."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server for Claude & Gemini Collaboration. Consult Gemini AI for a second opinion or collaborate on code development together.
consult_gemini - Get a Second OpinionAsk Gemini for advice, validation, or a different perspective on your work.
Use cases:
Validate your approach
Get code reviews
Ask for expert advice on technical problems
Get a different perspective on implementation choices
collaborate_on_code - Build TogetherCollaborate with Gemini to develop code from scratch through iterative dialogue.
Process:
Create and review PRD together
Decide on tech stack together
Generate and refine code together
Related MCP server: Claude Team MCP
npm install -g gemini-collaboration-mcpgit clone https://github.com/henry2craftman/gemini-collaboration-mcp.git
cd gemini-collaboration-mcp
npm install
npm run buildGet your API key from Google AI Studio
Add to your Claude Code MCP settings:
For npm installation:
{
"mcpServers": {
"gemini-collaboration": {
"command": "npx",
"args": ["-y", "gemini-collaboration-mcp"],
"env": {
"GEMINI_API_KEY": "your-gemini-api-key-here"
}
}
}
}For local installation:
{
"mcpServers": {
"gemini-collaboration": {
"command": "node",
"args": ["path/to/gemini-collaboration-mcp/dist/mcp-server.js"],
"env": {
"GEMINI_API_KEY": "your-gemini-api-key-here"
}
}
}
}Restart Claude Code to load the MCP server.
Once configured, you can use these commands in Claude Code:
"Gemini, what do you think about this approach?"
"Ask Gemini to review this code"
"Get Gemini's opinion on this implementation""Collaborate with Gemini to build a 3D dice game"
"Work with Gemini to create a calculator app"
"Develop a todo list app together with Gemini"This MCP server uses the AI Orchestration Framework to:
Chain AI model interactions
Manage context between Claude and Gemini
Execute multi-step collaborative workflows
Node.js 18 or higher
Gemini API key
Claude Code (Anthropic's official CLI)
# Install dependencies
npm install
# Build
npm run build
# Run MCP server directly
npm run mcp
# Run interactive chat
npm run chatMIT
Contributions are welcome! Please open an issue or submit a pull request.
For issues, questions, or suggestions, please open an issue on GitHub.
Collaborate with Gemini to develop code through iterative dialogue. Claude and Gemini will:
Create and review PRD together
Decide on tech stack together
Generate and refine code together Returns the final code after collaboration.
| Name | Required | Description | Default |
|---|---|---|---|
| request | Yes | What to build (e.g., "3D dice game", "calculator app") |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It outlines the collaborative steps (PRD creation, tech stack decision, code generation/refinement) but fails to disclose critical traits like whether this is a read-only or mutating operation, authentication requirements, rate limits, or what happens to intermediate outputs. The description is insufficient for a tool with complex multi-step behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized with three sentences that efficiently outline the collaboration process and outcome. It's front-loaded with the core purpose, though the numbered steps could be slightly more concise. Overall, it avoids unnecessary elaboration.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complex multi-step collaboration process, no annotations, and no output schema, the description is incomplete. It doesn't explain the format or structure of the returned 'final code', what happens during the iterative dialogue, or how errors or interruptions are handled. The description fails to compensate for the lack of structured metadata.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% with one parameter 'request' documented as 'What to build'. The description adds no additional parameter semantics beyond what the schema provides, maintaining the baseline score of 3 since the schema adequately covers the single parameter.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool enables collaboration with Gemini to develop code through iterative dialogue, specifying the verb 'collaborate' and resource 'code'. It distinguishes from the sibling 'consult_gemini' by emphasizing joint development rather than consultation, though it doesn't explicitly contrast them.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for code development projects requiring iterative collaboration, but provides no explicit guidance on when to use this tool versus 'consult_gemini' or other alternatives. It mentions the collaborative process but lacks clear exclusions or prerequisites.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Consult Gemini AI for a second opinion or advice. Use this when:
You need validation of your approach
You're uncertain about a decision
You want a different perspective
You need expert advice on a technical problem
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | The question or problem to consult Gemini about | |
| context | No | Additional context to help Gemini understand the situation (optional) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. While it mentions the tool's purpose and usage scenarios, it doesn't disclose important behavioral traits like whether this is a read-only operation, potential costs/rate limits, response format expectations, or any authentication requirements. The description is insufficient for a tool with zero annotation coverage.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured with a clear purpose statement followed by a bulleted list of usage scenarios. It's appropriately sized and front-loaded with the core purpose. The bullet points could potentially be more concise, but overall the structure is effective with minimal wasted space.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity of consulting an AI system, the lack of annotations, and no output schema, the description is incomplete. It doesn't address important contextual factors like response format, error conditions, rate limits, or what constitutes a successful consultation. For a tool that interacts with an external AI service, more behavioral context is needed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema has 100% description coverage, with both parameters ('query' and 'context') clearly documented in the schema itself. The tool description doesn't add any parameter-specific information beyond what's already in the schema, so it meets the baseline of 3 for high schema coverage without adding extra value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Consult Gemini AI for a second opinion or advice.' This specifies the action (consult) and resource (Gemini AI). However, it doesn't explicitly differentiate from the sibling tool 'collaborate_on_code' - both could involve AI assistance, so the distinction isn't articulated.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit usage guidelines with a bulleted list of four specific scenarios: validation of approach, uncertainty about decisions, different perspectives, and expert technical advice. This gives clear guidance on when to use this tool, though it doesn't mention when NOT to use it or explicitly contrast with the sibling tool.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
collaborate_on_codeconsult_geminiScored across 2 tools
The two tools have clearly distinct purposes: collaborate_on_code is for iterative code development through a structured collaboration process, while consult_gemini is for seeking validation, advice, or a second opinion on decisions or technical problems. There is no overlap in functionality or ambiguity between them.
Both tools follow a consistent verb_noun pattern: collaborate_on_code and consult_gemini. The naming is predictable and readable, with no deviations or mixed conventions, making it easy for agents to understand the action and target.
With only 2 tools, the server feels thin for its purpose of 'Gemini Collaboration,' which suggests a broader scope involving code development and consultation. While the tools are well-defined, the count is too low to provide comprehensive coverage for collaboration workflows, lacking tools for specific actions like reviewing code, managing iterations, or handling feedback loops.
The server has significant gaps in its tool surface for collaboration. It lacks tools for key operations such as reviewing or editing code independently, managing project states, or handling iterative feedback beyond the initial collaboration. This incomplete coverage will likely cause agent failures when trying to perform nuanced or multi-step collaboration tasks.
One identity across Claude Code, Codex, Cursor, Gemini, Windsurf: shared inbox and handoffs.
Agent personas for Claude. 16 tools, 13 personas, 3 workflows. Zero extra API cost. Free.
One MCP endpoint for Claude, GPT & Gemini: 100+ tools + no-code connectors + agent workers.
Build and supervise fleets of agents from Claude Code, Codex or Cursor. Connects over OAuth.
