setup-vault-types

type	skill
name	setup-vault-types
description	Interactive wizard to configure which document types your vault uses. Asks the user what kinds of notes they take (journals, books, meetings, clients, podcasts, etc.), enables the matching extractors, and scaffolds extractors for custom types. Run once after installing ai-brain-starter, and again whenever you add a new kind of doc. Do NOT use this to run extraction. Use /second-brain-mapping for that.
trigger	/setup-vault-types
argument-hint	[--add <typename> | --list | --remove <typename>]
tool_access	["Read","Write","Edit","Glob","Bash"]
policy_constraints	[{"rule":"Never delete an extractor file the user has customized","exception_handling":"If the user requests removal of a customized extractor, prompt for explicit confirmation before deleting"},{"rule":"Never enable an extractor whose target type the user did not confirm","exception_handling":"List the type and ask the user to confirm before enabling"},{"rule":"When scaffolding a new extractor, write to the extractors directory only","exception_handling":"Refuse to write outside the extractors path even if a path argument suggests it"}]
required_inputs	[{"name":"mode","type":"string","required":false,"description":"One of --add, --list, --remove. If omitted, runs interactive wizard mode."},{"name":"typename","type":"string","required":false,"description":"Extractor type slug. Required when mode is --add or --remove."}]
output_shape	{"format":"terminal-report","fields":{"enabled_types":"list of extractor type slugs now enabled in the vault","scaffolded_types":"list of new extractor files written, with their paths","summary_line":"one-line confirmation of what changed"}}

/setup-vault-types

Figure out which doc types belong in this vault, then wire the right extractors.

Why

/second-brain-mapping ships with 18 extractors. You probably don't need all of them. This wizard:

1. Asks what kinds of notes you take
2. Enables the matching extractors
3. Offers to scaffold new extractors for types we don't ship

No "start small" suggestion. You get all the capability for the types you have.

Steps

Step 1 — Detect existing types

Scan the vault for files already declaring type: in frontmatter:

cd "$(vault-root)"
grep -rh "^type:" --include="*.md" -I . 2>/dev/null \
  | sed 's/^type:\s*//' | sort | uniq -c | sort -rn | head -30

Tell the user: "Your vault already has these types declared: …"

Step 2 — Ask what they take notes on

Present this list, ask user to pick any/all that apply:

Core journaling & reflection:
  [ ] journal        — daily reflection / gratitude / mood
  [ ] daily_log      — task-log style (Roam/Capacities daily)
  [ ] goal           — OKRs, quarterly plans, vision docs

Reading & learning:
  [ ] book           — book notes, highlights, reviews
  [ ] article        — saved essays, blog posts, long reads
  [ ] concept        — evergreen concept notes (PKM backbone)

People & relationships:
  [ ] person         — CRM / contact notes
  [ ] meeting        — 1:1s, team syncs, call notes

Creative & publishing:
  [ ] writing_draft  — blog drafts, book chapters, newsletters
  [ ] talk           — speaking engagements, workshops given

Work & business:
  [ ] business       — pitches, memos, client docs, investor updates
  [ ] strategy       — strategic plans, frameworks, bets
  [ ] negotiation_prep — deal prep, BATNA docs
  [ ] company        — entity notes (past/current ventures)
  [ ] ai_chat        — saved AI conversations (Claude, GPT, etc.)
  [ ] playbook       — SOPs, step-by-step guides

Lifestyle:
  [ ] travel         — trip notes, restaurants, places visited

Assets & reference:
  [ ] asset          — brand files, logos, templates
  [ ] reference      — cheat sheets, quick-lookup docs

Custom:
  [+] Add your own type

Step 3 — Install selected extractors

For each checked type, symlink its extractor into the vault's scripts/extractors/ dir. Leave unchecked types uninstalled — no wasted files.