user-interview

name	user-interview
description	Use during morning routine Step 7.5, DM replies, the morning briefing, and the evening sweep to manage the profile-interview queue at state/profile-questions.md — pick a latent question, weave it into a natural reply, tick when answered, fallback-promote stale rows.
allowed-tools	["Bash(curl *)","Read"]

Profile Interview Queue Skill

The agent has a slow-pace queue of profile facts to learn (name, timezone, location, work, hobbies, …). Questions are NOT pushed at the user as cold scheduled DMs. They wait in latent state until a natural opportunity arrives (a topically-related inbound DM, or the morning briefing). This skill is how every callsite touches that queue.

Source-of-truth file

state/profile-questions.md — agent-internal markdown. Three top-level sections: ## Pending, ## In Progress, ## Answered. Never auto-injected into prompts; load only when this skill is in use.

curl -s http://localhost:8321/api/context/state/profile-questions

Pending row format

- [ ] (PRIO) <id> :: <target_path[ ## Section]> [:: match=<anchor>] :: <ask-hint in English> [<!-- last_attempted=YYYY-MM-DD -->]

Field	Meaning
PRIO	HIGH / MID / LOW — selection priority
<id>	snake_case unique. Doubles as the correlation id in any scheduled DM description (profile_interview:<id> — <hint>)
target_path	which user file the answer should land in, e.g. identity/profile.md
## Section	optional — narrows to a section within the target file
match=<anchor>	optional — bullet key (English, like Name, Timezone, Sleep, Working hours). Required when multiple rows share a section, or when setup pre-seeds the section with an unrelated bullet
ask-hint	English brief of WHAT to ask (agent-internal, Policy A). Render the actual DM per <output_language_policy> — this skill intentionally splits the two surfaces
last_attempted=...	optional inline HTML comment maintained by the evening sweep — selector deprioritises rows whose comment is < 7 days old

In Progress entry format

- <id> :: state=<state> :: since=<YYYY-MM-DD> [:: scheduled_at=<ISO>] [:: asked_at=<ISO>]

since= is the agent-day date the entry was first added (set by the morning routine when picking the question). Load-bearing: the evening sweep's latent-fallback promotion (Operation 5B) computes today − since to decide whether ≥ 3 days have elapsed without an opportunity. Without it the sweep cannot tell a day-1 latent row from a day-3 one. since= is preserved across state=latent → asked → resolved transitions; only the state field flips.

State machine:

Pending
  └── morning routine Step 7.5 ──▶ latent
                                     │
            ┌─ DM-handler topic match ┘
            │  / briefing piggyback
            ▼
          asked ──── user replies ────▶ resolved (Pending row [x])
            ▲
            └─ scheduled (fallback) ── fires DM ─┐
                ▲                                 │
                │ 3 days latent + active user    │
                └─────────────────────────────────┘

• The DM-handler queue-flip MUST gate on state=asked only. An unrelated 09:00 DM cannot close out a 14:00 question that has not been asked yet.

Answered entry format

Append-only log:

- [x] YYYY-MM-DD → <id> (<source>)

Sources:

• (DM) — user answered in chat (load-bearing: never untick)
• (import:<source>) — profile import migration (never untick)
• (reconciled:skeleton|morning|opportunity|fire-time|sweep) — heuristic / LLM closure

"Section is filled" check (Layers 2/3/5)

Before scheduling or asking, every callsite SHOULD verify the target slot is genuinely empty:

curl -s "http://localhost:8321/api/profile-questions/slot-filled?path=identity/profile.md&section=Identity&anchor=Name"
# → {"filled":true|false,"sectionPresent":true|false,"fileExists":true|false,...}

This wraps the canonical TS helper. Use it — do NOT re-derive the rule in prose. URL-encode section and anchor if they contain spaces.

Operation 1 — Pick a question (Morning Routine Step 7.5)

Morning-routine-only. The full skip gates (5 conditions), walk order, drift-tick recovery, today.md mirror line, and no-schedule rule are in the op-morning reference below.

{{> ref:op-morning }}

Operations 2 & 4 — DM-handler weave and answer capture

DM-handler-only. Operation 2 (latent opportunity check) runs AFTER the standard "Capture user info" block in message.received.dm.md / message.received.dm_first.md, BEFORE composing the reply; Operation 4 (answer capture) runs after Operation 2's opportunity check. The slot-filled pre-check, naturalness rules, two PATCHes, and the "one DM = one tick" rationale are in the op-dm-handler reference below.

{{> ref:op-dm-handler }}

Operation 3 — Latent piggyback (morning briefing composition)

Scheduled.dm-only. Mirror of Operation 2 but run inside the ## Morning briefing sub-flow of scheduled.dm.md. Slot-filled pre-check, domain-overlap judgment, and the two PATCHes are in the op-briefing reference below.

{{> ref:op-briefing }}

{{> ref:sweep-and-fallback }}

Today.md surfacing — Agent Notes flavor

A latent question is mirrored to today.md ## Agent Notes for visibility. This is informational only — Agent Plan keeps its strict HH:MM-+-schedule contract; latent rows do NOT belong there.

- Profile question (latent): <id> — wait for natural opportunity
- Profile question (asked HH:MM): <id>        # after Operation 2/3/6

The morning routine writes the (latent) line. The DM handler / morning briefing flips the parenthetical to (asked HH:MM) when the question is woven into a reply. The evening sweep's stale recovery removes the line entirely if the user did not reply within 24h.

Anchor convention — load-bearing

The heuristic anchor lookup matches against bullet keys written by the user-profile skill: English label keys, any-language values:

- Name: Alex
- Timezone: America/New_York
- Working hours: Weekdays 09:00–18:00
- Sleep: 23:00–07:00

Anchors in state/profile-questions.md MUST match these English keys — not the user's primary-language phrasing. If a future user-profile schema change introduces non-English keys, Layers 1–3 silently miss those bullets and the system degrades to Layer-4-only protection (≤ 24h staleness). Cross-reference: agent-assets/skills/user-profile/SKILL.md §"File schema" + setup.initial.md (canonical bullet examples).

Rules

• Never ask via cold standalone DM if a natural opportunity is plausible.
• Never weave a question that is unrelated to the current conversation topic — better to wait.
• Never ask twice the same agent-day.
• Never tick a row whose target section is genuinely placeholder-only.
• Never write to ## Pending from any callsite other than skeleton seeding, Layer 4 untick (sweep), and Phase 2 evening-review extension.
• Never ask the user identity-class confirmations (name, timezone, primary_language) twice — these have explicit setup paths; if they are still empty after setup, the queue may ask once.

API quick reference

# Read the queue
curl -s http://localhost:8321/api/context/state/profile-questions

# Slot-filled probe
curl -s "http://localhost:8321/api/profile-questions/slot-filled?path=identity/profile.md&section=Identity&anchor=Name"

# Section-level edit (queue file uses the standard context API)
curl -s -X PATCH http://localhost:8321/api/context/state/profile-questions \
  -H 'Content-Type: application/json' \
  -d '{"section": "in_progress", "mode": "replace", "content": "- name :: state=latent"}'

# Fallback DM scheduling (Operation 5B only) — see the canonical
# POST /api/schedule example in references/sweep-and-fallback.md Op 5B.
# `prompt` is REQUIRED (the wake-up session has NO memory); `description`
# carries the `profile_interview:<id>` prefix that triggers Operation 6.

The PATCH section argument is snake_case of the heading: pending, in_progress, answered. Read-before-write applies to every PATCH replace.