From rpi
Diagnose a bug iteratively: reproduce, find root cause, attempt up to 3 fixes, produce a diagnosis artifact. Use when user says 'why is X not working?', 'X is broken', 'after Y, I see Z fail', or reports broken behavior, even if they don't say 'diagnose'. Do NOT invoke for understanding working code (use rpi-research).
How this skill is triggered — by the user, by Claude, or both
Slash command
/rpi:rpi-diagnoseThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Iteratively diagnose and fix complex bugs through root-cause analysis. This combines research's interview flow with implement's execution loop, adding an autonomous iterate-until-fixed cycle.
Iteratively diagnose and fix complex bugs through root-cause analysis. This combines research's interview flow with implement's execution loop, adding an autonomous iterate-until-fixed cycle.
Accept a bug description, error message, or path to a failing test. Clarify expected vs actual behavior, reproduce, trace root cause, then attempt fixes autonomously (up to 3 attempts). Produce a diagnosis artifact in .rpi/diagnoses/ regardless of outcome.
When the bug is fixed, announce completion and suggest → /rpi:rpi-commit to commit the fix and diagnosis artifact. When escalation is needed, suggest → /rpi:rpi-plan (or /rpi:rpi-propose for architectural issues).
/rpi:rpi-commit to review and commit both the fix and the diagnosis artifact together.rpi/diagnoses/ containing bug report (expected/actual/reproduction), root cause with file:line references, investigation log (each attempt with hypothesis/change/result), and resolution status/rpi:rpi-plan; if the fix requires architectural changes, suggest /rpi:rpi-propose instead/rpi:rpi-plan is more valuable than a fragile hackGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub a-ngj/rpi --plugin rpi