collect-feedback

To collect writing feedback, the Skill loads your original drafts and final published Medium or LinkedIn posts, compares them to surface differences, and translates those edits into actionable style rules for future drafts.

Updating writing style memory involves comparing the final published post against the original draft to extract structural learnings, then safely persisting those actionable rules to memory/brand/writing_style.md to reduce future editing cycles.

Log post-publication feedback rounds by recording each analysis cycle in memory/brand/feedback-log.md, ensuring you maintain a persistent, traceable history of style learnings extracted from your published Medium and LinkedIn content.

Yes, you can extract actionable style rules by loading originals and finals from your draft and published posts, analyzing the edits made, and translating those structural differences into guidelines for future content creation.

No external dependencies are required. The feedback loop operates independently to enforce validation, handle errors, and safely update memory/brand/writing_style.md and feedback log files to prevent data loss during the style learning process.