Document Store Bootstrapper
Parent: Document & File Formats · Published reference · snapshot 2026-09-08 · skill content-ingestion-extraction/references/doc-store-bootstrapper.md
↓ Facts as markdownall context files
Turns a folder of documents into an "ideal doc store" with a numbered taxonomy, _meta/ indexes, archival policy, and resumable operator memory. Use for Google Drive engagement folders, customer KBs, p
These notes link each claim to its source. Some citations identify archived research reports that are no longer hosted on this site. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
Document Store Bootstrapper
- Turns a folder of documents into an "ideal doc store" with a numbered taxonomy, _meta/ indexes, archival policy, and resumable operator memory. Use for Google Drive engagement folders, customer KBs, project doc stores, and research dumps. For code repositories, use repo-bootstrapper instead. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#document-store-bootstrapper]
When to use this skill
- Organize a flat folder into a clean, numbered taxonomy [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
- Stand up a new engagement/customer/project folder to the standard [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
- Upgrade an existing partly-organized folder toward the ideal state [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
- Audit a folder and produce a gap report [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
- Generate or refresh _meta/ index, manifest, and policy files [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
- Resume an interrupted run from _meta/memory.md [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#when-to-use-this-skill]
Skill guidance
- Treat the target as an ideal-state doc store, not a minimally compliant one. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Infer taxonomy from the actual documents present - do not invent categories with no documents. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Preserve .gdoc, .gsheet, .gslides, .gform, .gjam shortcuts byte-for-byte (Drive links depend on exact filename). [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Never rename or move a load-bearing filename without explicit confirmation. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Prefer non-destructive organization: move into folders, do not delete. Send deletion candidates to 99 Archive/. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- For Drive-synced folders: use mv (not cp) so Drive sees a move event. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Use _meta/memory.md as resumable state. If interrupted, the next run reads it and continues from the first pending step. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
- Log the user's prompt and any clarifying answers to _meta/prompts.md before moving any files. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#skill-guidance]
Standard file manifest
- Every compliant doc store must have ALL of the following inside _meta/: [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#standard-file-manifest]
Standard taxonomy
- Numerically prefixed so folders sort predictably. Adapt labels to the domain; keep the numbering scheme and reserved slots. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#standard-taxonomy]
- Empty slots are permitted but must be documented in _meta/taxonomy.md with a rationale, or removed if unused. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#standard-taxonomy]
Naming and archive rules
- Preserve .gdoc, .gsheet, .gslides, .gform, .gjam shortcuts exactly. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Strip trailing whitespace on rename; never on move. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Date prefixes use YYYY-MM-DD . [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Never rename load-bearing artifacts (scripts referenced by tooling, Drive-linked shortcuts). [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Move to 99 Archive/ when ALL of: last modified > 180 days ago OR explicitly superseded, AND not currently referenced by an active initiative or open case. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Always archive files starting with Copy of , (1), or – copy. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
- Use 99 Archive/<year>/ subfolders when volume warrants. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#naming-and-archive-rules]
Execution phases
- Run in order. Phases 3–6 may run in parallel agents. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#execution-phases]
Audit checklist
- Run against path/_meta/ and the folder layout: [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
- _meta/memory.md - versioned format, status field, Done/Pending/Resumption sections, owner and target path. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
- _meta/taxonomy.md - every numbered top-level folder explained; empty folders justified or removed; "what goes where" mapping. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
- _meta/INDEX.md - one entry per folder; top 5 key docs listed; last-refreshed date in header. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
- _meta/manifest.json - valid JSON; one entry per non-trivial document (exclude .DS_Store, hidden files); each entry has path, kind, mtime, tags[]. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
- README.md - names the folder and purpose; links to _meta/INDEX.md and _meta/taxonomy.md; last-refreshed date. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#audit-checklist]
Bootstrap invocation
- > Bootstrap the folder at <absolute path> to the doc-store-bootstrapper standard. Follow the Execution Phases in ~/.claude/skills/doc-store-bootstrapper/SKILL.md. Capture my prompt to _meta/prompts.md before moving any files. Use _meta/memory.md for resumable state. [archived source: https://llms-explorer.com/sources/mdb-context-hub/doc-store-bootstrapper/#bootstrap-invocation]
Where this helps
- A shared Google Drive folder for a customer engagement has grown into a flat pile of files with no naming convention and needs to become navigable before a new team member joins. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Standing up a brand-new customer, project, or engagement folder from day one, so it starts in the standard taxonomy instead of drifting into chaos over months. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Auditing a partially organized folder to produce a gap report before a handoff, showing what's missing from the standard file manifest. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Any doc store that needs resumable operator memory — where whoever picks up the folder next (human or agent) needs a _meta/ index to reconstruct state without re-reading everything. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
Project ideas
- Build a script that scans an existing flat folder, infers a numbered taxonomy from the file types and names actually present, and proposes a migration plan before moving anything. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Build an audit tool that checks a folder tree against the standard file manifest and _meta/ index requirements and emits a gap report with specific missing files named. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Build a Drive-safe rename/move utility that preserves .gdoc/.gsheet/.gslides shortcuts byte-for-byte while normalizing date prefixes to YYYY-MM-DD across a folder. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Build a periodic sweep job that flags load-bearing filenames (referenced by other tooling) before an automated cleanup would otherwise rename or move them. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
Antipatterns
- Inventing taxonomy categories that sound tidy but have no actual documents behind them, instead of deriving the numbered scheme from what's really in the folder. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Deleting files during a cleanup pass instead of moving deletion candidates into 99 Archive/, which is the only non-destructive path the standard allows. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Renaming or moving a load-bearing filename (one referenced by scripts or Drive-linked shortcuts) without confirming nothing else depends on the exact old path. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Leaving taxonomy slots empty without documenting the rationale in _meta/taxonomy.md, which makes the empty slot look like an oversight to the next person who opens the folder. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
Known issues
- Google Drive shortcut files (.gdoc, .gsheet, .gslides, .gform, .gjam) are placeholders, not real files — renaming them incorrectly can break the link back to the live document. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- Bootstrapping assumes write access to reorganize the folder; a read-only or heavily permissioned Drive folder limits how much of the standard structure can actually be applied. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- A folder with many collaborators actively adding files during the bootstrap pass can drift out of sync with the taxonomy the moment the audit finishes. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
- The standard taxonomy and manifest are opinionated; a folder with a pre-existing, well-functioning convention may be better left alone than forced into the numbered scheme. [source] — AI-suggested, synthesized from this pack's existing facts — not extracted from a source document.
Children
- Google Drive Engagement Folders (frontier)
- Document Taxonomy Design (frontier)
- Index and Metadata Standards (frontier)
- Archival Policy (frontier)
Frontier under this node: Archival Policy, Document Taxonomy Design, Google Drive Engagement Folders, Index and Metadata Standards