Gaz MCP

gaz-mcpis an MCP server for read-only MySQL/PostgreSQL access, Jenkins administration with configuration history, OpenRouter image generation, and diagnosis of running Go processes throughpprofandexpvar. It also ships the independentdiagnosticsGo module for adding those endpoints to target services.

- The Go version declared ingo.mod(currently Go1.26.0) to build from source, or a prebuilt binary.
- A MySQL/PostgreSQL server and/or Jenkins instance reachable from the host running the MCP.

# Latest release curl -fsSL https://raw.githubusercontent.com/jcastilloa/gaz-mcp/master/scripts/install.sh | sh # Specific version curl -fsSL https://raw.githubusercontent.com/jcastilloa/gaz-mcp/master/scripts/install.sh | VERSION=vX.Y.Z sh

Product-facing agent skills are distributed underskills/, separately from the.codex/skillsused to develop this repository:

GAZ_MCP_DIR=/path/to/gaz-mcp CONSUMER_PROJECT=/path/to/project-using-gaz-mcp mkdir -p "$CONSUMER_PROJECT/.codex/skills" cp -R "$GAZ_MCP_DIR/skills/gaz-mcp-db" "$CONSUMER_PROJECT/.codex/skills/" cp -R "$GAZ_MCP_DIR/skills/gaz-mcp-jenkins" "$CONSUMER_PROJECT/.codex/skills/" cp -R "$GAZ_MCP_DIR/skills/gaz-mcp-diagnostics" "$CONSUMER_PROJECT/.codex/skills/"

See theskills catalogfor project/global paths in Claude Code, Codex, OpenCode, and Cursor, plus the boundary between product and contributor skills.

service: transport: stdio version: 0.4.0 # SQL environments (MySQL + PostgreSQL) environments: dev1: engine: mysql host: 127.0.0.1 port: 3306 user: readonly_user password: your-password analytics: engine: postgres host: 127.0.0.1 port: 5432 user: postgres password: your-password # Jenkins environments jenkins: production: url: https://jenkins.example.com user: admin api_key: "${JENKINS_PROD_API_KEY}" # Jenkins API token or password timeout: 30s insecure: false staging: url: https://jenkins-staging.example.com user: admin api_key: "${JENKINS_STAGING_API_KEY}" timeout: 30s insecure: true # allow self-signed TLS # Jenkins configuration history (SQLite, pure Go) snapshot: enabled: true db_path: ~/.config/gaz-mcp/jenkins_history.db max_versions: 50 # positive retention limit per object auto_prune: true # OpenRouter image generation. models is a closed allowlist: image_generate # rejects every model not declared here. image_generation: api_key: "${OPENROUTER_API_KEY}" base_url: https://openrouter.ai/api/v1 timeout: 2m models: - openai/gpt-image-1 - google/gemini-2.5-flash-image

Seeconfig.sample.yamlfor the complete sample, including the OpenAI provider settings used by the application.

cp config.sample.yaml config.yaml # Edit the environments, jenkins, snapshot, and image_generation sections, then render secret placeholders.

go run ./cmd/server/ --transport stdio # Or with the built binary: ./gaz-mcp --transport stdio

{ "mcpServers": { "gaz-mcp": { "command": "/absolute/path/to/gaz-mcp", "args": ["--transport", "stdio"] } } }

[mcp_servers.gaz-mcp] command = "/absolute/path/to/gaz-mcp" args = ["--transport", "stdio"] startup_timeout_sec = 20.0

codex mcp add gaz-mcp -- /absolute/path/to/gaz-mcp --transport stdio

Follow the prompts: project or global, namegaz-mcp, typelocal, then command/absolute/path/to/gaz-mcp --transport stdio.

Always use an absolute binary path and stdio transport.

The root document contains installation, configuration, client setup, architecture, and development. The complete operational references live in the focused guides below; none of the tool reference is intentionally duplicated here.

The service-side module is released independently using module-compatible tags such asdiagnostics/v0.1.0; MCP server release tags (v0.4.x) are not module versions forgithub.com/jcastilloa/gaz-mcp/diagnostics.

To move the service endpoints from/debug/...to a custom prefix, usediagnostics.WithBasePath("/internal/diagnostics"); then passhttp://service:6060/internal/diagnosticsas the diagnostic tooltarget.

gaz-mcp/ ├── cmd/server/ # Entry point ├── diagnostics/ # Independent service-side instrumentation Go module ├── skills/ # Product-facing agent skills distributed to gaz-mcp users ├── mcp/ │ ├── application/ │ │ ├── diagnostics/ # Go process diagnosis use cases │ │ ├── image/ # Image-generation use cases │ │ ├── jenkins/ # Jenkins use cases + NoopSnapshotRepository │ │ └── sql/ # SQL use case (read-only enforcement) │ └── domain/ │ ├── diagnostics/ # Diagnostic target and profile-store ports │ ├── image/ # Image-generation repository port │ ├── jenkins/ # Repository + SnapshotRepository ports │ └── sql/ # Repository port ├── platform/ │ ├── config/ # Viper config reader │ ├── di/ # Dependency injection container │ └── mcp/ │ ├── commands/ # Cobra runner + tool wiring │ ├── diagnostics/ # HTTP pprof/expvar adapter + profile store │ ├── image/ # OpenRouter image API adapter │ ├── jenkins/ # gojenkins infrastructure adapter │ ├── server/ # MCP server wrapper │ ├── snapshot/ # SQLite snapshot repository │ ├── sql/ # MySQL + PostgreSQL adapters │ └── tools/ # MCP tool definitions └── shared/ ├── ai/domain/ # AI provider contracts └── config/domain/ # Configuration contracts

Dependency rule:platform → shared + mcp/application + mcp/domain. Never reverse that direction.

go build ./... # Build all packages go vet ./... # Static analysis go test ./... # All tests go test ./mcp/application/jenkins/... -v # Jenkins service unit tests go test ./platform/mcp/snapshot/... -v # SQLite snapshot integration tests go test ./mcp/application/diagnostics/... -v # Diagnostic service tests go test ./platform/mcp/diagnostics/... -v # HTTP adapter and profile-store tests (cd diagnostics && go test ./...) # Service-side instrumentation module

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

The 1Password MCP server creates a bridge that allows MCP clients such as Codex and Kiro to manage your 1Password Environments with secure authorization prompts.

This is the 1st, easiest, and cheapest PPT, slides, presentation AI generation MCP Server in the world.

Persistent memory for any AI assistant. Zero token cost until recall. Stores memories in local SQLite, ranks by 6-factor scoring, returns results 79% smaller than JSON. Works with Claude, ChatGPT, Grok, Cursor, Windsurf, and any MCP client.

A MCP server that enables AI assistants to interact with Anki, the spaced repetition flashcard application.

Enables LLM clients to interact with macOS applications through AppleScript. Built using the @beyondbetter/bb-mcp-server library, this server provides safe, controlled execution of predefined scripts with optional support for arbitrary script execution.

An MCP server for WordPress plugin audits

Turn your AI assistant into a digital marketing hub that creates, organizes, and analyzes links and QR Codes on demand.

Connect AI clients to Cal.com scheduling through the Model Context Protocol using the hosted server at mcp.cal.com or a local instance.

Sync Calendars, Scheduling Links, AI Executive Scheduling Assistant, Unified Calendar