Agent skill

smith-tools

Tool configurations for IDEs, MCP integrations, and development tools. Use when configuring IDE settings, MCP tools, or pytest. Covers conditional tool activation, configuration hierarchy, and synchronization patterns.

View SKILL.md on GitHub Repository
Stars 1
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/tianjianjiang/smith/tree/main/smith-tools

SKILL.md

Tool Configurations

  • Scope: Configuration standards for development tools, IDEs, and MCP integrations
  • Load if: Writing/editing IDE config files (.vscode/, .kiro/, .cursor/) OR configuring MCP tools
  • Prerequisites: @smith-principles/SKILL.md, @smith-standards/SKILL.md

CRITICAL: Tool Configuration (Primacy Zone)

  • Check codebase and local docs BEFORE using MCP tools
  • Use MCP tools only when task specifically requires them
  • Activate tools conditionally, not by default
  • Mandating Serena MCP for all plans (only multi-session)
  • Requiring Context7 for all libraries (only unfamiliar external)
  • Loading MCP tools unconditionally

This document defines configuration standards for development tools, IDEs, and external integrations.

IDE & Extension Settings

  • Global Settings: $HOME/Library/Application Support/[Code|Cursor|Kiro]/User/settings.json
  • Workspace Settings: .vscode/settings.json, .cursor/settings.json, .kiro/settings/
  • Claude Code: Same configuration across all IDEs, follow skill standards
  • Consistency: Synchronized extension preferences across all environments

Pytest Configuration (MANDATORY)

  • Log Level: log_cli_level = "WARNING" in both pytest.ini and pyproject.toml
  • Purpose: Suppress INFO/DEBUG logs from external libraries during test execution
  • Exception: Use DEBUG level only when actively debugging specific issues

For detailed pytest execution patterns: See @smith-python/SKILL.md#action-recency-zone

MCP Tool Integration (Optional)

MCP (Model Context Protocol) tools provide enhanced capabilities for specific scenarios. Use them conditionally when their specific functionality is needed.

These tools are NOT required for standard development work.

Available MCP Tools

Serena MCP - Session Persistence:

  • Purpose: Persist plans and context across sessions
  • Use for: Complex multi-session tasks, plan recovery, session restarts
  • Configuration: .kiro/settings/mcp.json or IDE settings
  • Avoid: Single-session tasks, simple bug fixes

Context7 - External Library Documentation:

  • Purpose: Fetch documentation for external libraries
  • Use for: Unfamiliar NPM/PyPI packages, external API references
  • Configuration: MCP settings
  • Avoid: Libraries already in codebase, standard library usage

Fetch/WebFetch - Web Content Retrieval:

  • Purpose: Retrieve web content from URLs
  • Use for: User-provided URLs, web documentation explicitly requested
  • Configuration: Use WebFetch tool if available
  • Avoid: Content available locally, speculative browsing

For MCP server configuration in Kiro: See @smith-ctx-kiro/SKILL.md#critical-serena-mcp-is-mandatory

MCP Server Lifecycle

Installation:

shell
claude mcp add --transport stdio --scope user server-name -- command args

Scopes: --scope local (default, ~/.claude.json), --scope project (.mcp.json in repo root), --scope user (~/.claude.json)

Discovery:

shell
claude mcp list

Removal:

shell
claude mcp remove server-name

Debugging failures:

  • Check logs: ~/.claude/logs/mcp-*.log
  • Startup timeout: relaunch Claude Code with a higher MCP_TIMEOUT env var
  • Restart: remove + re-add the server
  • Config reload: restart Claude Code session after changes to .mcp.json or settings

Common issues:

  • Server not found: verify command path is absolute
  • Permission denied: check executable permissions
  • Startup timeout: increase MCP_TIMEOUT for slow servers

Configuration Hierarchy

  • $HOME/.smith/ - Global personal standards (symlinked to projects)
  • AGENTS.md - Project-level agent instructions
  • .vscode/, .cursor/, .kiro/settings/ - IDE-specific configs
  • MCP tool configs - Serena, Context7, etc.

Synchronization Points

  • IDE Settings: Automatically sync from master documents
  • Project Files: Reference personal standards instead of duplicating
  • Optional Tools: Configure MCP tools conditionally based on task needs
  • Tool Extensions: Apply these standards in all generated content

Update Workflow

  1. Master Update: Modify authoritative documents
  2. Automatic Sync: Run synchronization script to update all tool configurations
  3. Validation: Verify consistency across all tools and configurations
  4. Documentation: Update project-specific documentation with references
  • @smith-ctx-claude/SKILL.md - Claude Code patterns
  • @smith-prompts/SKILL.md - Structured output patterns
  • @smith-python/SKILL.md - Virtual environment, pytest patterns
  • @smith-ide/SKILL.md - IDE-specific path syntax

ACTION (Recency Zone)

Before using MCP tools:

  1. Check codebase and local docs first
  2. Use MCP only when local info insufficient
  3. Configure tools conditionally per task

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

tianjianjiang/smith

smith-style

File naming, path standards, and conventional commits. Use when naming files, creating branches, writing commit messages, or setting up new projects. Covers underscore vs hyphen conventions, commit format, and branch naming patterns.

1 0
Explore
tianjianjiang/smith

smith-stacks

Stacked pull request workflows for large features. Use when creating stacked PRs, managing dependent PRs, or rebasing after parent merges. Covers stack creation, merge order, and squash merge handling.

1 0
Explore
tianjianjiang/smith

smith-python

Python development with uv, pytest, ruff, and type hints. Use when writing Python code, running tests, managing Python packages, or working with virtual environments. Covers import organization, type hints, pytest patterns, and environment variables.

1 0
Explore
tianjianjiang/smith

smith-principles

Fundamental coding principles (DRY, KISS, YAGNI, SOLID, HHH). Use when starting any development task, evaluating implementation approaches, or reviewing code quality. Always active as foundation for all development decisions.

1 0
Explore
tianjianjiang/smith

smith-nuxt

Nuxt 3 development patterns including auto-import stubbing for tests, environment variable conventions, and middleware testing. Use when working with Nuxt projects, testing Nuxt components/middleware, or configuring Nuxt environment variables.

1 0
Explore
tianjianjiang/smith

smith-plan

Plan tracking protocol (portable). Progress tracking with checkboxes, iteration workflow, completion/blocker signals. Use when executing multi-step plans, tracking task progress, or working from plan files. IMPORTANT - Always update the plan file after completing tasks.

1 0
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results