{"id":"zeromicro-mcp-zero","name":"mcp-zero","homepage":null,"repo_url":"https://github.com/zeromicro/mcp-zero","category":"devtools","subcategories":[],"tags":["mcp","code-generation","go-zero","goctl","go","grpc","rest","developer-tools","claude-desktop","microservices"],"what_it_does":"mcp-zero is an MCP (Model Context Protocol) server (stdio-based) that helps scaffold and generate go-zero projects and components using go-zero’s goctl CLI. It exposes multiple MCP tools to generate API services, RPC services, API code from specs, database models, configs/templates, analyze existing projects, query go-zero docs, and validate inputs.","use_cases":["Scaffolding new go-zero REST API services from prompts","Generating gRPC/RPC services from provided protobuf content","Generating go-zero API code from .api specification files","Producing database models from MySQL/PostgreSQL/MongoDB schemas or DDL","Creating go-zero configuration files and common templates (middleware/error handler/deploy artifacts)","Validating API spec/proto/config content before running code generation","Analyzing an existing go-zero project structure and dependencies","Answering go-zero documentation/migration questions via an MCP tool"],"not_for":["Production security-critical API endpoints (it’s a local developer tool, not a deployed service)","Handling highly sensitive credentials end-to-end without careful secret management and filesystem access controls","Automated code generation in restricted environments where writing to arbitrary output directories is not allowed","Teams that require OAuth/user-based access control to tool calls"],"best_when":"You want an agent-driven workflow to generate go-zero boilerplate and code locally, using Claude Desktop (or another MCP-compatible client) to call these MCP tools.","avoid_when":"You cannot control where generated code lands (output_dir/output_path) or you cannot safely provide database/proto/spec content and connection strings to the tool.","alternatives":["go-zero/goctl directly (CLI-based scaffolding without MCP)","Other MCP or codegen tools integrated with go-zero (if available)","General-purpose code generators/templating without go-zero-specific validation (lower correctness)"],"af_score":61.5,"security_score":31.8,"reliability_score":32.5,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:25:59.791428+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":true,"sdk_languages":["Go"],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["None described (local MCP tool via stdio)","Environment-based configuration (GOCTL_PATH)"],"oauth":false,"scopes":false,"notes":"README describes running an MCP server locally and configuring the client to start it. No user authentication/authorization mechanisms are described for tool calls."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No pricing model described; likely open-source tooling."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":61.5,"security_score":31.8,"reliability_score":32.5,"mcp_server_quality":75.0,"documentation_accuracy":70.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":95.0,"rate_limit_clarity":0.0,"tls_enforcement":30.0,"auth_strength":30.0,"scope_granularity":10.0,"dependency_hygiene":55.0,"secret_handling":40.0,"security_notes":"Security features are only lightly described. The MCP server is stdio/local (so TLS is not a primary control), and no authentication/authorization is described. The README mentions 'safe credential handling with environment variable substitution' and logging/metrics, but provides no concrete details about redaction, secure storage, or how credentials are handled in prompts vs. logs. The tools accept database connection strings and arbitrary content; without strong input/output controls, there is risk of credential exposure and unintended filesystem writes.","uptime_documented":0.0,"version_stability":45.0,"breaking_changes_history":50.0,"error_recovery":35.0,"idempotency_support":"false","idempotency_notes":"No idempotency guarantees are mentioned. Generators write files; repeated runs may overwrite or create changes depending on tool behavior.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Tool parameters include filesystem paths (output_dir/output_path/api_file/project_dir). Agents should be cautious about overwriting files and validate paths.","Database model generation requires database connection strings/DDL; agents should avoid leaking secrets and ensure network/database accessibility.","Some tool inputs are large blobs (proto_content, content). Agents should respect client/tool message size limits.","Port/style defaults exist but may not match existing project conventions; agents should confirm style and output paths."]}}