Skills before: 38 | Skills after: 27 active + 13 archived Shared rules created: 7 | Scripts created: 3 Model Assumptions added: 27/27 active skills New skills created: 2 (vault-quarterly, vault-half-review) — pending /skill-creator refinement
| Skill | Lines | Reason |
|---|---|---|
| rfc | 54 | replaced by write-an-rfc |
| session-summary | 206 | replaced by vault-session |
| daily-report | 191 | replaced by vault-daily |
| weekly-report | 196 | replaced by vault-weekly |
| monthly-report | 86 | replaced by vault-monthly |
| quarterly-report | 67 | replaced by vault-quarterly (new) |
| half-review | 74 | replaced by vault-half-review (new) |
| discord-prd | 61 | replaced by write-a-prd |
| map-button-routes | 158 | button migration done |
| validate-button-routes | 73 | button migration done |
| split-button-pr | 313 | button migration done |
| build-move-table | 78 | uikit-native migration done |
| move-native-file | 94 | uikit-native migration done |
| Rule | Skills using it |
|---|---|
image-diff-workflow.md |
experiment-visual-diff, migrate-button, screenshot-migrated-buttons |
sozluk-integration.md |
do-work, prd-to-tasks, qa, vault-session, vault-daily, vault-context, vault-weekly, write-an-rfc, write-a-prd, sozluk, harness-audit |
vault-paths.md |
all vault-* skills, do-work, prd-to-tasks, qa, sozluk, bootstrap-*, classify-ticket |
linking-rules.md |
vault-session, vault-daily, vault-weekly, vault-monthly, vault-quarterly, vault-half-review, debug-writeup, investigation |
reporting-backfill.md |
vault-daily, vault-weekly, vault-monthly, vault-quarterly, vault-half-review |
progress-frontmatter.md |
do-work, prd-to-tasks, qa, vault-session, vault-context |
ios-simulator-navigation.md |
migrate-button, screenshot-migrated-buttons, experiment-visual-diff |
| Script | Purpose |
|---|---|
dssim-compare |
DSSIM score for full images or cropped elements |
resolve-project |
Find vault project for current repo |
check-backfill |
Find missing vault reports in a date range |
model: claude-opus-4-6— pins each skill to its validated modeleffort: high|medium— optimizes token usage per skill complexityargument-hint— where applicable, for autocomplete UX
Every skill now has a ## Model Assumptions section documenting:
- What the instruction encodes as an assumption
- Whether it's still load-bearing
- How to test/stress-test it
From Anthropic's harness design post: "every component in a harness encodes an assumption about what the model can't do on its own, and those assumptions are worth stress testing."
The harness-audit skill was updated to make assumption inference mandatory (not optional). Future audits will compare inferred assumptions against documented ones and flag obsolete workarounds.
- Run
/skill-creatorfor vault-quarterly (rough draft exists) - Run
/skill-creatorfor vault-half-review (rough draft exists)