Recover Neotoma SQLite Database
SkillDatabases & dataCheck Neotoma SQLite integrity and run .recover when the DB is corrupt (malformed disk image, btree errors). Use when MCP/API fails with SQLite corruption.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Recover Neotoma SQLite Database skill
What this skill tells your AI
The instructions your AI receives, as published by markmhendrickson/neotoma in skills/recover-sqlite-database/SKILL.md and read by ahel’s review.
When to use
Neotoma errors such as database disk image is malformed, failed PRAGMA integrity_check, or btreeInitPage / SQLITE_CORRUPT on the local SQLite file (neotoma.db or neotoma.prod.db under NEOTOMA_DATA_DIR).
Preconditions
sqlite3CLI onPATH(macOS usually has it).- Stop Neotoma (MCP stdio processes,
neotoma api, anything using the DB) before--recoveror before swapping files.
Script (primary)
From the Neotoma install location:
# Check only (exit 0 = ok, 2 = corrupt)
npx neotoma storage recover-db
npx neotoma storage recover-db --env production
# Write recovered copy (does not replace live DB)
npx neotoma storage recover-db --env production --recover
Output is a sibling file like neotoma.prod.recovered-<timestamp>.db. Verify it reports PRAGMA integrity_check: ok, then manually archive the live .db/-wal/-shm and cp the recovered file to the live name.
After swap
- Inspect
lost_and_foundin the recovered file if present; re-ingest important rows viastore_structuredusing normal schemas. - Prefer
NEOTOMA_DATA_DIRoutside iCloud-synced folders to reduce recurrence.
Do not
- Run
--recoverwhile Neotoma still holds the database open. - Auto-delete the corrupt file until a good recovered copy is verified and backed up.
Signals
- GitHub stars
- 32
- Forks
- 3
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
recover-sqlite-database- Source
- github.com/markmhendrickson/neotoma