From dunk
Internal format specification for the DLN system. Only relevant when preloaded by the dln-sync agent via the skills frontmatter field. Never activated by user prompts. Defines the re-anchor payload compression template that dln-sync uses to convert raw Notion page-body read-backs into compact structured summaries for DLN teaching skills.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dunk:dln-compressThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill defines the compression format for converting raw Notion page content into a re-anchor payload. Follow these rules exactly when compressing read-back data.
This skill defines the compression format for converting raw Notion page content into a re-anchor payload. Follow these rules exactly when compressing read-back data.
After reading back Notion page content, produce EXACTLY this structure. Be concise — compress lists to key terms, not full sentences. The teaching skill needs just enough to re-orient, not the full raw content.
## Re-anchor
### Knowledge State
- Concepts: [comma-separated, each with (mastered)/(partial)/(not-mastered) status]
- Syllabus: [N/M covered (X mastered, Y partial, Z not-mastered) | Uncovered: topic1, topic2, ... OR "no syllabus"]
- Chains: [comma-separated abbreviated descriptions, each with mastery status]
- Factors: [comma-separated, each with mastery status]
- Compressed Model: [current model text, verbatim if short, summarized if >3 sentences]
- Weakness Queue: [top 2 items with severity, or "empty"]
- Calibration: [mean confidence gap, or "insufficient data"]
- Load: [current batch size, last overload session or "none"]
- Engagement: [momentum=positive/neutral/negative, struggles=N]
- Open Questions: [comma-separated list]
### Current Session
- Plan: [one-line summary of session plan]
- Progress: [bullet list of completed items with outcomes]
- Adjustments: [any plan changes, or "none"]
### Status
- Write: [success/failed]
- Failed writes: [list of what failed, or "none"]
- Review: [not due / due today / N days overdue / just completed — recall N%]
- Next Review: [date]
- Review Interval: [N days]
- Next on plan: [what the teaching skill should do next]
npx claudepluginhub luqmannurhakimbazman/ashford --plugin dunkGenerates learning notes from conversations: Classic mode for concise recaps, Mastery for detailed rules, misconceptions, and Core Loop code history. Supports Korean/English with beginner-friendly terms.
Reviews conversation, extracts learnings, and persists to auto-memory or vault. Run after substantial work sessions to avoid knowledge rot.
Formats responses with conclusion-first structure, concrete evidence, risk disclosure, and actionable next steps for efficient communication, task reports, planning, and writing feedback.