{"id":"gudastudio-geminimcp","name":"geminimcp","homepage":"https://code.guda.studio","repo_url":"https://github.com/GuDaStudio/geminimcp","category":"devtools","subcategories":[],"tags":["mcp","claude-code","gemini","python","developer-tools","ai-coding"],"what_it_does":"geminimcp is an MCP server that exposes Google Gemini CLI functionality through the Model Context Protocol so Claude Code (or other MCP clients) can invoke a `gemini` tool via stdio transport and use session-based interactions.","use_cases":["Use Claude Code as the front-end while leveraging Gemini CLI capabilities for coding tasks","Generate front-end/UI code prototypes (HTML/CSS/React/Vue) via MCP tool calls","Multi-turn agent workflows using SESSION_ID for continuity"],"not_for":["Production systems needing guaranteed reliability/uptime guarantees","Use cases requiring fine-grained API governance (scopes, tenant isolation, centralized policy enforcement) without additional surrounding infrastructure","Cases needing documented model/provider authentication and secret management within the MCP server itself"],"best_when":"You already have Gemini CLI configured locally and want an MCP bridge for agent tooling (e.g., Claude Code) using stdio in a developer environment.","avoid_when":"You need a publicly hosted service with strong documented security controls, or you require clear, enforced rate limiting and structured error semantics documented by the MCP server.","alternatives":["Use Gemini CLI directly from the agent (without MCP)","Other Gemini-focused MCP servers (if available) or generic MCP wrappers around model APIs","Direct API integration to Gemini provider endpoints (if you can manage auth and tooling yourself)"],"af_score":53.5,"security_score":27.8,"reliability_score":22.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T13:34:01.322997+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["Local Gemini CLI configuration (implied) used by the MCP server"],"oauth":false,"scopes":false,"notes":"README does not describe explicit auth handled by the MCP server (e.g., API keys, OAuth) beyond relying on Gemini CLI configuration."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No pricing information for the MCP server itself is provided; Gemini CLI/model usage would typically incur model/provider costs depending on your Gemini setup."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":53.5,"security_score":27.8,"reliability_score":22.5,"mcp_server_quality":70.0,"documentation_accuracy":55.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":75.0,"rate_limit_clarity":5.0,"tls_enforcement":20.0,"auth_strength":30.0,"scope_granularity":10.0,"dependency_hygiene":55.0,"secret_handling":30.0,"security_notes":"Likely runs locally over stdio (no TLS applicable). README does not describe authentication, scopes, or secret-handling practices; it also does not document how Gemini credentials are stored/forwarded. For security, ensure Gemini CLI credentials are protected on the host and restrict who can run the MCP process.","uptime_documented":0.0,"version_stability":40.0,"breaking_changes_history":20.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":"No idempotency guarantees documented; `gemini` calls appear to be non-idempotent conversational operations.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Prompt/usage guidance provided in README is extensive but not enforceable by the MCP server itself—agents must follow it manually.","Context length limitations are mentioned in the recommended agent prompt, but not in MCP tool error responses.","Transport is stdio; deployments must ensure the MCP client runs the process correctly and allows the tool."]}}