AndyShaman/gemini-webapi-mcp — MCP server for Google Gemini — free image generation, editing & chat via browser cookies.

Description and use

**Brief Summary (2‑3 sentences)**
AndyShaman/gemini‑webapi‑mcp is a Python‑based MCP (Model Context Protocol) server that proxies Google Gemini’s image generation, editing, and chat capabilities without requiring API keys—authentication is handled via browser cookies. It provides a simple, standards‑based endpoint that AI agents can call to access Gemini’s multimodal functions, making it easy to integrate Gemini into tool‑use workflows. With recent commits, 34 stars, and a growing fork count, the project is mature enough for pilot deployments.

**Value**

• **Zero‑key access** – Developers can leverage Gemini’s powerful multimodal models without managing Google Cloud credentials, lowering cost and operational overhead.
• **Standardized interface** – By exposing Gemini through the Model Context Protocol, the server fits naturally into existing AI‑assistant architectures that already consume MCP endpoints, enabling plug‑and‑play tool integration.
• **Extensible backend** – Built in Python, the codebase is easy to extend (e.g., adding caching, rate‑limiting, or custom preprocessing) while keeping the core Gemini interaction unchanged.

**Practical Adoption Path**

1. **Clone & configure** – Pull the repository, set the required environment variables (Gemini cookie values), and run the provided Dockerfile or uvicorn command.
2. **Connect your agent** – Point your MCP‑compatible AI assistant (e.g., LangChain, AutoGPT, or a custom orchestrator) to the server’s /v1/chat/completions and /v1/images/generations endpoints.
3. **Test & iterate** – Use the built‑in Swagger UI or simple curl requests to verify image generation and chat responses, then add any domain‑specific wrappers or authentication layers.
4. **Scale** – Deploy the server behind a load balancer or Kubernetes service, optionally enabling TLS and rate‑limiting for production traffic.

**Production Readiness**

• **Activity & community** – Last updated on 2026‑06‑23, with steady commit history, 34 GitHub stars, and 19 forks, indicating active interest.
• **Maturity** – The codebase follows common Python web‑service patterns (FastAPI/Starlette), includes OpenAPI docs, and has clear entry points for CLI/SDK usage.
• **Risks** – The license, security posture, and long‑term maintainer commitment still need a final review, but no major metadata or dependency issues are evident.

Overall, the project is a strong OSS candidate for a serious pilot, offering a low‑friction way to bring Gemini’s capabilities into production AI pipelines.

Problem

AndyShaman/gemini-webapi-mcp helps connect AI assistants to real tools and data through a standard protocol.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

MCP server for Google Gemini — free image generation, editing & chat via browser cookies. No API keys needed.

Source