菜单
Guides the user through selecting and configuring an MCP (Model Context Protocol) server for a project. Recommends servers based on the stack, then walks through the just mcp-add wizard. Invoked when the user asks to set up MCP, add a tool integration, or configure external service access for AI agents.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Structures, writes, and reviews Terraform infrastructure code. Covers module layout, remote state, workspace strategy, variable and secrets handling, CI plan/apply pipeline, naming conventions, and multi-region deployment patterns (provider aliases, per-region state, failover strategies), while delegating shared risk classification, version guards, and rollback contract to `terraform-risk-playbook`. Invoked when the user asks to write Terraform, set up infrastructure as code, or review IaC.
Creates or updates AWS-oriented Terraform module structure and implementation scaffolding. Focuses on file layout, variable/output contracts, provider and version constraints, tagging standards, safe composition patterns, and address-stability guardrails.
Designs and implements Terraform module tests for AWS-targeted modules. Covers terraform-native tests, validation gates, fixture composition, and risk-focused assertions for regressions, address migration, and unsafe changes.
Designs AWS-targeted Terraform infrastructure plans before implementation. Focuses on service selection, module boundaries, state/backends, environment separation, security/compliance guardrails, acceptance criteria, and explicit risk/validation assumptions.
Diagnoses Terraform/OpenTofu risk before generating changes. Defines the response contract, failure-mode routing, version/runtime guards, validation commands, and rollback expectations for modules, tests, CI, and state work.
Uses lightweight AWS stack emulation for faster local Terraform iteration. Defines what can be safely validated locally, compatibility caveats, and the handoff boundary to real-AWS verification and reviewed plan/apply flows.
| name | configure-mcp |
| description | Guides the user through selecting and configuring an MCP (Model Context Protocol) server for a project. Recommends servers based on the stack, then walks through the just mcp-add wizard. Invoked when the user asks to set up MCP, add a tool integration, or configure external service access for AI agents. |
| version | 1.0.0 |
| tags | ["agentic","mcp","setup"] |
| resources | [] |
| vendor_support | {"claude":"native","opencode":"native","copilot":"prompt-inject","codex":"prompt-inject","gemini":"prompt-inject"} |
Guide the user through selecting and configuring an MCP server for their project.
For GitHub issue planning workflows, MCP is the preferred path for issue-backed
plan persistence. If MCP is unavailable or misconfigured, use gh CLI as the
primary fallback.
Ask the user:
Based on the stack, recommend from the canonical short-list:
| Server | When to use it |
|---|---|
github | GitHub repos, issues, PRs, CI status |
filesystem | Direct file system access within a bounded path |
postgres | Query and inspect a PostgreSQL database |
context7 | Live documentation lookup for libraries and frameworks |
playwright | Browser automation and end-to-end testing |
Explain which servers apply to the user's project and why.
Tell the user to run the interactive wizard:
just mcp-add /path/to/project
Walk them through the prompts:
github)stdio for local executables, http for remote serversnpx -y @modelcontextprotocol/server-github)GITHUB_TOKEN).mcp.json
6. Optionally sync to .gemini/settings.jsonAfter the wizard completes, verify with:
just mcp-list /path/to/project
This should show the newly configured server in the table.
Remind the user: environment variables referenced in the MCP config (e.g. ${GITHUB_TOKEN}) must be set in the shell before starting the AI tool. They are not stored in .mcp.json — only the variable references are.
Add them to the project's .env or the system environment as appropriate.