inbox-setup

---
name: inbox-setup
description: Use when the user wants to set up email triage for the first time: a one-time interview builds the taxonomy, voice, and preference records inbox-triage needs.
category: Productivity
version: 1.0.0
tools: []
---

# Inbox-Setup — Email Triage Onboarding

Paired with the companion **inbox-triage** skill: this is a one-time (or occasional re-run) interview that builds the structured Knowledge Base entries inbox-triage searches on every run — email taxonomy, reply patterns, evaluation framework, rate card, blocklist, tracker, and a triage log. The two skills' record structures must match exactly.

Run once, or re-run when business, pricing, or priorities change. Interview the user about their email patterns, business context, reply style, and priorities, then produce the structured records inbox-triage needs to process the inbox effectively.

## Invocation Triggers

"set up my inbox," "configure inbox triage," "set up my email system," "configure email triage," "build my email knowledge base," "initialize email management," "set up inbox triage," "onboard email triage."

## Conduct Discipline

**Do NOT generate all records at once.** Walk through the 8 sections one at a time. Each section commits its record(s) before moving on — return the finished record in chat for the Knowledge Base as soon as a section closes, so a dropped-off interview still leaves a usable partial setup.

Grill-me discipline applies throughout: **one question per turn**, never bundled, even across section boundaries. **"Why I'm asking" on every question.** **Forcing format where possible** — multi-choice beats open-ended. **Dependency-ordered** — Q2 depends on Q1, downstream sections depend on upstream.

## Knowledge Base Records To Produce

| Record | Purpose | Required? |
|---|---|---|
| Email Taxonomy | Classification system + report preferences | **Yes** |
| Email Patterns | Reply voice, tone, templates, hard rules | **Yes** |
| Evaluation Framework | Decision tree for opportunity emails | Only if user receives pitches/opportunities |
| Rate Card | Pricing, terms, negotiation posture | Only if user has pricing |
| Blocklist | Auto-skip senders + learned decline patterns | **Yes** (seeded, grows over time) |
| Tracker | Active follow-ups, overdue items, deadlines | **Yes** (starts mostly empty) |
| Triage Log | Per-run log entries | **Yes** (starts empty) |

## Stop Condition (Full Interview)

~25-31 questions total across the 8 sections (depending on skip-logic). Hard ceiling: 35 questions including all sub-clarifications. Section 4 (Evaluation Framework) is skipped entirely when Section 1 surfaced no opportunity-email category, dropping the total by 6 questions and the rate-card record. After Section 8's confirmation, intake is closed — **never re-open it**. To change preferences later, the user re-runs the skill, which detects existing records and asks per-record: replace / merge / skip. The grill-me one-at-a-time rule applies across section boundaries — do not batch questions moving from one section to the next.

## Section 1: The Big Picture

Six grill-me questions, one at a time:

- **What do you do?** Role and business in 1-2 sentences. *Why: context shapes what email patterns to expect — a solo creator's inbox looks nothing like an enterprise PM's.*
- **What dominates your inbox?** Top 1-2: sales pitches / client work / internal team / newsletters / customer support / financial / other. *Why: dominant categories drive the taxonomy.*
- **Rough volume split** — e.g. "60% business inquiries, 20% ops, 20% noise." *Why: tells me where to focus triage effort.*
- **Which email address(es) should triage cover?** *Why: if multiple, per-address taxonomies get set up.*
- **Run frequency:** once daily / 2x daily / 3x daily / on-demand only? *Why: drives the default search window in triage (9h overlap for 2x/day).*
- **Anyone helping manage email** — assistant, VA, team — or solo? *Why: persona handling differs for delegated inboxes.*

Build a mental model; don't produce records yet. Note whether opportunity emails are a category (drives Section 4 skip-logic).

## Section 2: Email Categories

Propose 5-7 categories based on Section 1 — pre-recommend a subset, not the whole menu: New Opportunities, Active Conversations, Action Required, Financial, Important/Personal, Informational, Ignore/Low Priority.

Three forcing questions: **Does this proposed taxonomy match your inbox reality — yes / mostly / no?** (if "no," redo the taxonomy before anything else makes sense). **Missing categories?** (skip if none — missing categories produce uncategorized emails downstream). **Which category takes the MOST time per email?** (that's where draft-reply effort needs to focus most).

Produce the Email Taxonomy record: categories, signals for each (trigger phrases / sender patterns / subject markers), and default actions per category.

## Section 3: Reply Style & Voice

Six grill-me questions: **Register** — formal / casual / in-between? **Three communication pet peeves** — phrases you hate, openings you avoid (treated as forbidden tokens in drafts). **Phrases or sign-offs you always use** (your voice fingerprints). **Different persona for different contexts** — e.g. assistant replies as you? **Typical reply length** — one-liner / short paragraph / longer? **Hard rules** — never X / always Y (e.g. never emojis, always reply within 24h) — enforced as non-negotiable in every draft.

**The critical highest-quality input:** ask the user to paste 3-5 real sent emails. Self-description of voice is unreliable — real samples are the best signal. Analyze them for opening phrases, sign-offs, length distribution, and register markers to supplement everything above. If the user runs a business, also ask about media kits, rate sheets, standard pitches, and repeated replies.

Produce the Email Patterns record: tone description with do/don't examples, persona rules, templates, signatures, hard rules.

## Section 4: Evaluation Framework (Conditional)

**Skip-logic:** only run this section if Section 1 surfaced opportunity emails as a meaningful category; otherwise jump straight to Section 5.

Six grill-me questions: **First thing you check when pitched something** — your gut filter (the top of the decision tree). **Three instant deal-breakers** (become PASS-auto signals). **Three things that make you immediately interested** (become TAKE-IT signals). **Standard pricing/terms** — or "no fixed pricing" if you negotiate every time (drives whether a Rate Card gets produced). **Negotiation posture:** firm / flexible / depends on context (drives draft tone on counter-offers). **VIP senders or organizations that always get engagement** (bypasses normal PASS filters).

Produce the Evaluation Framework record (decision tree + recommendation categories + VIP list) and the Rate Card record if pricing exists.

## Section 5: Blocklist & Patterns

Three grill-me questions: **Senders or domains to always skip** (auto-blocklist saves the most time per run). **Patterns in emails you always delete** — unsubscribe links from specific marketers, recruiter cold outreach, newsletters (patterns let triage auto-skip variants without exact-match maintenance). **Specific companies/recruiters/newsletters wasting time** (seeds the blocklist; triage adds more as overrides accumulate).

Produce the Blocklist record (auto-maintained by triage thereafter).

## Section 6: Current State

Three grill-me questions: **Active threads you're tracking** (become tracker entries so triage knows existing context). **Overdue replies** — anything you should have responded to but haven't (flagged as priority every run until resolved). **Time-sensitive items with deadlines** (enforced and surfaced as overdue at the right time).

Produce the Tracker record: active follow-ups, overdue section, resolved section (empty), update log (empty).

## Section 7: Report Preferences

Three grill-me questions: **Delivery format** — email draft to self / chat summary only? **Detail level** — 30-second scan / detailed breakdown / both? **Anything always shown first** — e.g. overdue payments, VIP messages (custom "top-of-report" rules).

Save these into the Email Taxonomy record under a "Report Preferences" section.

## Section 8: Confirmation & Handoff

List every record created with a one-sentence summary, then: "Your triage system is ready. Run the inbox-triage skill to process your inbox. First runs need oversight — the system learns from your edits and overrides." Remind the user to re-run this setup anytime business, pricing, or priorities change.

## Privacy Boundary

**Never persist passwords, full account numbers, SSNs, or other sensitive credentials in any record.** If the user volunteers such info during the interview, acknowledge it but don't store it — note `[stored separately by user]` in its place.

## Re-Run Behavior

Detect existing records; for each, ask per-record: replace / merge / skip. Walk only the sections whose records the user chose to update; skip sections whose records the user kept.

## Error Handling

| Situation | Behavior |
|---|---|
| No prior setup exists | Proceed with the full 8-section interview |
| User refuses to share samples | Use self-description; flag that voice calibration may need iteration |
| User says "skip this" mid-interview | Honor it; flag the gap as `[needs follow-up]` |
| Sensitive info volunteered | Acknowledge but don't persist; note `[stored separately by user]` |
| Re-run on existing setup | Detect existing records; ask per-record: replace, merge, skip |
| User has no pricing / opportunities | Skip Section 4 entirely; don't produce empty records |

## Anti-Patterns To Reject

- Generating all records at once instead of walking through sections
- Asking all questions in one batch
- Hardcoded provider references (Gmail-only thinking)
- Persisting sensitive credentials
- Skipping the "why this question matters" explanation
- Skipping the sample-emails ask for voice (it's the highest-quality input)
- Overwriting existing records without consent on re-run
- Forcing a Rate Card or Evaluation Framework record when they don't apply