Stats
Actions
Tags
From agentic-harness
Run a kaizen retrospective on an agentic-harness run and file tool-improvement feedback upstream (invokes the harness-feedback skill).
How this command is triggered — by the user, by Claude, or both
Slash command
/agentic-harness:feedback [optional: a quick note on what to give feedback about]The summary Claude sees in its command listing — used to decide when to auto-load this command
Run a kaizen / postmortem retrospective on the most recent agentic-harness work and, with my explicit approval, file tool-improvement feedback as a GitHub issue on the upstream `MrBogomips/agentic-harness` repository. Invoke the **harness-feedback** skill and follow it end to end: establish the run context, walk the retrospective, draft in the standard feedback format, redact project-identifying details, pass the completeness gate, and create the issue only after I approve the exact body. If `gh` is unavailable or unauthenticated, give me a prefilled new-issue URL instead. Starting note (...
Run a kaizen / postmortem retrospective on the most recent agentic-harness work and, with my
explicit approval, file tool-improvement feedback as a GitHub issue on the upstream
MrBogomips/agentic-harness repository.
Invoke the harness-feedback skill and follow it end to end: establish the run context, walk the
retrospective, draft in the standard feedback format, redact project-identifying details, pass the
completeness gate, and create the issue only after I approve the exact body. If gh is unavailable
or unauthenticated, give me a prefilled new-issue URL instead.
Starting note (optional): $ARGUMENTS
npx claudepluginhub mrbogomips/agentic-harness --plugin agentic-harness/feedbackSubmits feedback or issues for Ralph Specum plugin by creating GitHub issues via gh CLI if available, otherwise provides manual submission link and asks for message if none given.
/feedbackCaptures feedback on KARIMO or Claude Code operations with auto-detection of issue complexity, generates rules for simple cases or launches adaptive investigation for complex ones.
/feedbackIncorporates feedback into PR/FAQ document (prfaq.tex), redrafting affected sections with impact analysis. Supports direct text or batch from meeting-summary.md files.
/feedbackInteractively collects feedback on successes, corrections, agent issues, workflow problems, or suggestions, then submits to Supabase for learning system.