Vault Maintenance Recommendations
Recommendations are not authorization. Apply separately after review.
| Recommendation | Action type | Paths | Authority risk | Reason | Approval/status |
|---|---|---|---|---|---|
| Mark the baseline assessment as completed in QA/TODO | cross-link/status | 07 QA/Questions.md Q-024; 03 Context/CURRENT-TODO.md 0b | low | first full-vault synthesis now exists, but journal approval remains open | recommended; not applied |
| Link Baseline 001 from synthesis README and Current Pickup | cross-link | Weekly Synthesis README; Current Pickup | low | makes the first run discoverable without changing story priority | recommended; not applied |
| Correct stale Orzai journal wording | status wording | 02 Story/Characters/Orzai.md | low | character note says proposed while profile/registry say working visual development | recommended; not applied |
Audit and retire American-equivalent from compiled story | terminology cleanup | Samuel, Timeline, Great War component/event, treatment | significant if done blindly | rules prohibit real-world analogies in compiled story | requires approved invented functional description/name; preserve sessions |
| Add a workflow-to-authority status taxonomy | configuration | Rules or new QA reference; synthesis config | low–moderate | repository uses many valid workflow statuses that an automated pass could misread as story authority | recommended design task |
| Create relative chronology anchor table | new QA/story-support artifact | Timeline support file | moderate | exposes incompatibilities before exact dates are chosen | BUILD NOW; do not alter established durations |
| Add evidence-node template or extend Story Unit schema | workflow improvement | Story Loop templates | low | evidence recurs as setup/custody/authentication/reveal/consequence | recommended after baseline review |
| Consolidate duplicate open-question wording by cross-reference | maintenance | Open Questions, QA Questions, Current TODO | moderate | repeated signals are useful but can diverge | merge only indexes; preserve original sessions and question IDs |
| Record journal candidate limitations in any future gallery/index | registry/public boundary | journal candidate manifest and future public material | low | prevents attractive images becoming false authority | already documented locally; carry forward |
| Preserve all superseded chronology and Samuel-war material historically | archive guardrail | dated sessions, QA contradictions | high if removed | it records development and explicit supersession | no deletion or rewrite |
Promote only after author approval
No story promotion is recommended from this run. The assessment identifies gaps and candidate work only. In particular, do not promote any answer for primary status, hierarchy lock, Luminai capability, Witness identity, sabotage mechanism, terminal transfer, or final choice.
Merge or cross-link
After approval, link this baseline from the Weekly Synthesis README, Current Pickup, and Q-024. Cross-link repeated questions rather than deleting their appearances in context, QA, Story Units, and brainstorm packets.
Rename or archive candidates
No current candidate requires archival deletion. Journal images remain correctly outside approved references. The non-canon Great War Story Miner sample remains a useful workflow test and must keep its non-canon status.
TODO, QA, registry, and index updates
Recommended after author review: mark the synthesis portion of TODO 0b complete while leaving journal approval incomplete; mark Q-024 resolved with mode/date/modules; add WS-012 or equivalent terminology drift to QA if the author wants a tracked cleanup campaign.
Stale or superseded material to preserve historically
Preserve Samuel's former conquest/rampage model, former Luminai presentation rule, old Witness placements, and former development-style decision only in dated history and explicit supersession records. Do not let them re-enter current compiled summaries.