Stats
Actions
Tags
From spp
Mentors humans to complete coding tasks themselves for healthy SPP ratio. Provides goal overviews, file pointers, test instructions, guidance, and code reviews without AI commit attribution.
How this skill is triggered — by the user, by Claude, or both
Slash command
/spp:spp-help-human-codeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Guide a human through completing a coding task themselves. Use this skill when the SPP ratio is unhealthy (too much AI-written code) and the human needs to write code to improve their ratio. Be a mentor for your human friend, help them stay sharp and learn programming.
Guide a human through completing a coding task themselves. Use this skill when the SPP ratio is unhealthy (too much AI-written code) and the human needs to write code to improve their ratio. Be a mentor for your human friend, help them stay sharp and learn programming.
spp status shows the human ratio is below targetDescribe the high level goal: Explain what we are trying to accomplish at a high level.
Offer code pointers to relevant files: Show the user which files they need to modify, and which lines, and what they need to do there. You may include code snippets on trickier algorithms or parts that require more arcane syntax knowledge.
Provide test instructions: Tell the user how they can test their change and verify correctness.
Ask if the user wants more help: If they want more help, provide guidance including code snippets or other in more depth hand holding.
Offer to review the user's code: After they are done writing code, review it using standard best practices for code review.
Important: Do not commit user's code with 'Co-Authored-By: Claude' If the user asks you to commit their code, make sure not to include the 'Co-Authored-By: Claude' note in the commit message. Otherwise, SPP will track the commit as written by Claude and not the human.
npx claudepluginhub mlolson/claude-spp --plugin sppPrompts developers to explain AI-generated code or plans via rubber duck questioning to verify comprehension and prevent rubber-stamping.
Narrates decisions, tradeoffs, and reasoning in plain language while building, helping users learn by working alongside Claude. Activates on requests for teaching or mentoring.
Manages AI pair programming sessions with driver/navigator roles, TDD, real-time code review, and quality verification. Invoke when collaborating on code with role-based workflows.