Frameworks for effective mentoring and knowledge transfer. Use for 1:1 meetings, pair programming, onboarding, teaching technical concepts, and developing junior engineers.
Provides frameworks for effective developer mentoring, including Crawl-Walk-Run progression, Socratic questioning, and trust-building. Use when onboarding engineers, running 1:1s, pair programming, or teaching technical concepts to junior developers.
/plugin marketplace add melodic-software/claude-code-plugins/plugin install soft-skills@melodic-softwareThis skill is limited to using the following tools:
references/one-on-one-structure.mdreferences/pair-programming-guide.mdThis skill provides frameworks for effective mentoring, knowledge transfer, and developing other engineers.
A framework for teaching new skills progressively:
| Phase | Mentor Role | Mentee Role | Duration |
|---|---|---|---|
| Crawl | Do, they observe | Watch and ask questions | Until they understand the "what" |
| Walk | Guide heavily | They try, you correct | Until they can do it with help |
| Run | Provide guardrails | They lead, you advise | Ongoing with decreasing support |
Crawl: You review PRs together, thinking aloud about what you look for, why things matter, what makes good/bad code.
Walk: They do the review, you watch. You ask questions: "What about this section?" You course-correct in real-time.
Run: They review independently. You spot-check occasionally and discuss any disagreements. They come to you with edge cases.
Key principle: Stay in each phase long enough. Rushing to "Run" creates gaps.
Instead of giving answers, ask questions that lead to understanding:
| Instead of... | Ask... |
|---|---|
| "Use a hash map here" | "What data structure would give us O(1) lookups?" |
| "You need to handle null" | "What happens if this value is null?" |
| "That's inefficient" | "What's the time complexity here? Could we do better?" |
| "Don't do it that way" | "What are the trade-offs of this approach?" |
When NOT to use Socratic questioning:
Trust is the foundation of effective mentoring:
Show genuine interest in their goals
Create psychological safety
Maintain confidentiality
Be consistent and reliable
Acknowledge when they teach you
People learn differently. Adapt your approach:
| Style | Signs | Approach |
|---|---|---|
| Visual | Asks for diagrams, draws things out | Use whiteboarding, architecture diagrams, code walkthroughs |
| Auditory | Learns from discussion, podcasts | Talk through concepts, think-aloud, verbal explanations |
| Kinesthetic | Prefers hands-on practice | Pair programming, experiments, building things |
| Reading/Writing | Prefers documentation | Point to docs, have them write summaries |
Most people are a mix. Start with all approaches, then observe what clicks.
Pair programming is a powerful mentoring tool when done well. See references/pair-programming-guide.md for detailed guidance.
Effective 1:1s are the backbone of mentoring. See references/one-on-one-structure.md for detailed templates.
❌ Grabbing the keyboard when they struggle ✅ Ask guiding questions, let them try
❌ "You know what a REST API is, right?" ✅ "What's your experience with REST APIs?"
❌ Explaining everything about microservices at once ✅ Focus on what they need now, save rest for later
❌ Only discussing technical work ✅ Check in on how they're doing personally
❌ "I'll just fix this, it's faster" ✅ "Let's fix this together so you see how"
Track mentee development over time:
references/pair-programming-guide.md - Communication during pairingreferences/one-on-one-structure.md - 1:1 meeting frameworks/soft-skills:write-1on1-agenda command - Generate 1:1 agendasfeedback-conversations skill - Giving developmental feedbackprofessional-communication skill - General communication patternsDate: 2025-12-26 Model: claude-opus-4-5-20251101
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.