--- title: "Vault Maintenance" slug: "vault-maintenance" category: "knowledge-vaults" tags: ["knowledge-vaults", "maintenance", "review", "dataview", "quality-gates", "workflow"] status: "stable" last_updated: 2026-08-29 summary: "Weekly, monthly, and quarterly review protocols, the Dataview gates each one walks, and the maintenance budget that decides how much architecture a vault can carry." related: ["[[knowledge-vaults/vault-audit]]", "[[knowledge-vaults/atomic-notes]]", "[[knowledge-vaults/linking-and-tags]]", "[[knowledge-vaults/decision-journals]]", "[[knowledge-vaults/vault-evolution]]", "[[knowledge-vaults/semantic-audit]]", "[[tooling/obsidian-plugins]]"] --- ## Overview A vault is maintained or it is abandoned. The maintenance is a set of scheduled passes over a fixed list of queries, not a vague intention to tidy up. This page gives the cadences, the gates each one walks, and the rule that decides how much structure the vault can afford in the first place. ## Size the architecture to the maintenance budget, not the ambition Ask how many minutes per week the owner will actually spend, then design to that number. A vault that needs 90 minutes a week from someone who will give 20 is abandoned inside two months, and an abandoned vault returns nothing. At a low budget, cut hard and say what was cut: five folders instead of thirteen, three note types instead of nine, the weekly review only, no quarterly audit, no spaced repetition. The system survives on weekly hygiene; everything else is an upgrade the habit earns. Cutting layers is covered as a legitimate evolution move in [[knowledge-vaults/vault-evolution]]. ## Run the weekly review as a fixed 30-minute checklist The weekly review is the load-bearing ritual. It walks the same queries every time, in the same order. 1. **Inbox to zero.** Every capture becomes an atom, a source note, or a deletion. 2. **Orphan sweep.** Link it or delete it, no third outcome. 3. **Source backfill.** Every unsourced claim gets a source link, or `confidence: low` and a self-derived flag. 4. **Stale seedlings.** Atoms 30 days old and still seedling: promote or delete. 5. **Atomicity candidates.** Notes past the word threshold: split or trim. 6. **Promotion pass.** Would I defend this in writing? If yes, promote. 7. **Decision check-ins.** Any decision past its expected resolution date gets its outcome and lesson written now. Wire the queries into the review note so the pass is one document walked top to bottom rather than seven searches. That is the difference between a review that happens and one that does not. ## Build the gates as saved queries, not as habits Every rule in this category should have a query that finds its violations. A rule with no query is a preference. ```dataview TABLE file.mtime AS "Modified", confidence FROM "20-Atoms" WHERE (!sources OR length(sources) = 0) AND type = "atom" SORT file.mtime ASC ``` The core set: orphan detector, unsourced claims, stale seedlings, over-length atoms, open questions, stale verification on decision-relevant claims, unresolved contradictions past their aging threshold, decisions due for review, and source notes with zero atoms extracted. Keep them in one dashboard note. The plugin setup is covered in [[tooling/obsidian-plugins]]. ## Reserve the monthly review for drift, not hygiene Hygiene is weekly. The monthly pass asks whether the vault is still pointed at the right thing. - Scope drift: is the work still the work described when the vault was designed? - Domain tags: are they still mutually exclusive, and should any merge or split? - Stale verification: re-verify or downgrade every decision-relevant claim past its threshold. - Contradictions: any closer to resolution, or ready to abandon? - Source diet: which authors and formats dominate, and what does that skew? - Synthesis prompt: what is the vault now close to being able to write? Run the deterministic auditor before this review and clear the criticals first, per [[knowledge-vaults/vault-audit]]. A strategic review over a vault with unresolved criticals is reasoning on unverified material. ## Use the quarterly audit to test the vault, not to read it Quarterly, sample 20 atoms at random and read them. Still useful, still correct, still atomic? A random sample finds decay that no query catches, because the notes that rot are the ones nobody opens. Also quarterly: the decision pattern review by class from [[knowledge-vaults/decision-journals]], a restore test of the backup, and the semantic pass in [[knowledge-vaults/semantic-audit]] if the vault is large enough to warrant it. ## Track the trend, not the snapshot Log every audit run and every review, one line each. The value is in the direction of travel: orphans down, unsourced atoms up, three decisions overdue for a second week. A rising violation count is the early signal that the weekly review has quietly stopped happening, and it shows up in the log weeks before it shows up as a vault nobody opens. Trends also tell you when a folder or note type has outgrown its design, which is the trigger for the evolution moves in [[knowledge-vaults/vault-evolution]]. ## Related - [[knowledge-vaults/vault-audit]] - [[knowledge-vaults/atomic-notes]] - [[knowledge-vaults/linking-and-tags]] - [[knowledge-vaults/decision-journals]] - [[knowledge-vaults/vault-evolution]] - [[knowledge-vaults/semantic-audit]] - [[tooling/obsidian-plugins]]