@windyroad/itil 0.26.0 → 0.27.0-preview.296

package/scripts/test/skill-md-dual-tolerant-coverage-rfc-002-t3.bats

@@ -0,0 +1,374 @@
+#!/usr/bin/env bats
+
+# @rfc RFC-002 T3 — Bats fixture audit + dual-tolerant assertions
+# @adr ADR-031 (Problem-ticket directory layout — per-state subdirs)
+# @adr ADR-051 (load-bearing-from-the-start — each SKILL-prescribed
+#   enumeration pipeline ships with a behavioural enforcement test
+#   exercised against per-state-layout synthetic fixtures, not later
+#   by graceful drift discovery during the T5 migration cutover)
+# @adr ADR-052 (behavioural-bats default — these tests run the actual
+#   shell pipelines the SKILL.md sites prescribe against synthetic
+#   fixtures and assert observable enumeration. They do NOT structurally
+#   grep SKILL.md prose for the dual-pattern string, which would be
+#   P081-class structural-test-disguised-as-behavioural and was
+#   explicitly excluded from T2 per architect finding 3)
+# @adr ADR-014 (single-purpose: one mechanical contract — the
+#   SKILL-prescribed pipelines compose with per-state-layout fixtures)
+# @adr ADR-060 (Phase 1 Slice 5 forward-dogfood — T3 commit grain)
+# @problem P069 (driving — flat layout unskimmable; the migration this
+#   contract guards is the relief)
+# @problem P081 (no structural-grep on SKILL.md content — this test
+#   is the behavioural alternative)
+# @jtbd JTBD-001 (extended scope — multi-commit RFC-grain coordinated
+#   change governance; T3 is one of 11 RFC-002 sub-tasks; behavioural
+#   coverage is how per-edit governance stays trustworthy across the
+#   migration window)
+# @jtbd JTBD-006 (work-backlog-AFK — dual-tolerant pipelines preserve
+#   AFK-loop continuity during the T2-to-T6 migration window; without
+#   this contract, mid-migration AFK iterations silently miss tickets
+#   in the un-migrated layout half)
+# @jtbd JTBD-008 (decompose-fix-into-coordinated-changes — RFC-002 T3
+#   is the load-bearing test artefact for the coordinated-change
+#   sub-workstream; visible as an RFC-002-T3 entity rather than
+#   diffusing across 14 existing files per JTBD review)
+# @jtbd JTBD-101 (extend-the-suite — adopter projects consuming
+#   @windyroad/itil at the T2-shipped state must enumerate correctly
+#   against their flat-layout tickets AND post-auto-migration per-state
+#   tickets; T3 proves both halves)
+#
+# Contract: every SKILL.md call site updated in T2 (commit `0795e91`,
+# 14 SKILL.md surfaces) prescribes a shell pipeline of canonical shape:
+#
+#   ls docs/problems/*.<state>.md docs/problems/<state>/*.md 2>/dev/null
+#   ls docs/problems/*.md         docs/problems/*/*.md       2>/dev/null
+#   ls docs/problems/<ID>-*.md    docs/problems/*/<ID>-*.md  2>/dev/null
+#
+# T2's `dual-tolerant-glob-rfc-002-t2.bats` exercises the canonical
+# pattern shapes generically. T3 extends that coverage to the
+# end-to-end SKILL-prescribed PIPELINES — the next-ID compute pipeline
+# (`ls X Y | sed | grep -oE | sort -n | tail -1`), the multi-state
+# union form (4-pathspec for open + known-error backlog scan), the
+# verifying-state filter as run-retro Step 4a dispatches it, and the
+# brace-expansion ID + state-set form report-upstream uses. Each test
+# runs the pipeline against three synthetic fixture shapes (flat-only,
+# per-state-only, mixed) and asserts observable enumeration.
+#
+# T6 (post-T5 verification) drops the flat-layout half. This test
+# updates at T6 to single-pattern (per-state only), NOT removed — the
+# contract narrows but the behavioural enforcement remains.
+#
+# CONTRACT NOTE: when one half of the dual-pattern has zero matches
+# in the current fixture (single-layout fixtures), `ls X Y 2>/dev/null`
+# exits nonzero — the unmatched literal pathname propagates to ls's
+# argv and `2>/dev/null` only suppresses the stderr noise, not the
+# exit code. SKILL.md call sites MUST treat STDOUT emptiness as the
+# canonical "no tickets" signal, NOT exit code zero. Test assertions
+# probe stdout content via `run` (which absorbs the exit code into
+# `$status`); `$status` is asserted only in the empty-fixture and
+# missing-ID cases where nonzero exit is the intended contract.
+
+setup() {
+  REPO_ROOT="$(mktemp -d)"
+  cd "$REPO_ROOT"
+  mkdir -p docs/problems
+}
+
+teardown() {
+  cd /
+  rm -rf "$REPO_ROOT"
+}
+
+# ── fixture builders ─────────────────────────────────────────────────────────
+
+build_flat_layout() {
+  cat > docs/problems/100-foo.open.md <<'EOF'
+# Problem 100: Foo
+**Status**: Open
+**WSJF**: 5.0
+EOF
+  cat > docs/problems/101-bar.known-error.md <<'EOF'
+# Problem 101: Bar
+**Status**: Known Error
+**WSJF**: 4.0
+EOF
+  cat > docs/problems/102-baz.verifying.md <<'EOF'
+# Problem 102: Baz
+**Status**: Verification Pending
+EOF
+  cat > docs/problems/103-qux.parked.md <<'EOF'
+# Problem 103: Qux
+**Status**: Parked
+EOF
+  cat > docs/problems/104-quux.closed.md <<'EOF'
+# Problem 104: Quux
+**Status**: Closed
+EOF
+}
+
+build_per_state_layout() {
+  mkdir -p docs/problems/open docs/problems/known-error docs/problems/verifying docs/problems/parked docs/problems/closed
+  cat > docs/problems/open/200-foo2.md <<'EOF'
+# Problem 200: Foo2
+**Status**: Open
+**WSJF**: 6.0
+EOF
+  cat > docs/problems/known-error/201-bar2.md <<'EOF'
+# Problem 201: Bar2
+**Status**: Known Error
+**WSJF**: 3.5
+EOF
+  cat > docs/problems/verifying/202-baz2.md <<'EOF'
+# Problem 202: Baz2
+**Status**: Verification Pending
+EOF
+  cat > docs/problems/parked/203-qux2.md <<'EOF'
+# Problem 203: Qux2
+**Status**: Parked
+EOF
+  cat > docs/problems/closed/204-quux2.md <<'EOF'
+# Problem 204: Quux2
+**Status**: Closed
+EOF
+}
+
+build_mixed_layout() {
+  build_flat_layout
+  build_per_state_layout
+}
+
+# ── Pipeline 1: Next-ID compute (manage-problem Step 3 + capture-problem Step 3)
+#
+# SKILL.md prescribes the recursive local_max formula:
+#   ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null \
+#     | sed 's|.*/||' \
+#     | grep -oE '^[0-9]+' \
+#     | sort -n | tail -1
+#
+# Architect finding 2 (T2): the recursive enumeration MUST contribute
+# tickets from BOTH layouts to max-ID, otherwise a per-state ticket at
+# ID 204 is invisible to a flat-only-enumerating capture-problem and
+# the next ID re-allocates an already-taken slot. T3 exercises the
+# pipeline against per-state-only AND mixed fixtures to prove the
+# dual-pathspec composes with the downstream sed/grep/sort pipeline.
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "next-ID pipeline: flat-only fixture yields max ID 104" {
+  build_flat_layout
+  run bash -c "ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1"
+  [ "$output" = "104" ]
+}
+
+@test "next-ID pipeline: per-state-only fixture yields max ID 204" {
+  build_per_state_layout
+  run bash -c "ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1"
+  [ "$output" = "204" ]
+}
+
+@test "next-ID pipeline: mixed fixture yields max ID 204 (recursive enumeration spans both layouts)" {
+  # Architect finding 2: a per-state ticket at ID 204 MUST contribute
+  # to max-ID even when flat-layout 104 also exists. Drop the per-state
+  # half of the dual-pathspec and this test fails — capture-problem
+  # would re-allocate ID 105 instead of advancing to 205.
+  build_mixed_layout
+  run bash -c "ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1"
+  [ "$output" = "204" ]
+}
+
+@test "next-ID pipeline: empty fixture yields empty result" {
+  run bash -c "ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1"
+  [ -z "$output" ]
+}
+
+# ── Pipeline 2: Open + known-error multi-state union
+#   (work-problems Step 1, list-problems live scan)
+#
+# SKILL.md prescribes the 4-pathspec form:
+#   ls docs/problems/*.open.md docs/problems/*.known-error.md \
+#      docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null
+#
+# This is wider than T2's single-state filter — it unions two states
+# across two layouts in one ls invocation. The prove-out shape: the
+# union enumerates open + known-error from BOTH layouts and excludes
+# verifying / parked / closed from BOTH layouts.
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "open+known-error union: per-state-only fixture enumerates 200 and 201" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null'
+  [[ "$output" == *"open/200-foo2.md"* ]]
+  [[ "$output" == *"known-error/201-bar2.md"* ]]
+}
+
+@test "open+known-error union: per-state-only fixture excludes verifying/parked/closed" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null'
+  [[ "$output" != *"202-baz2"* ]]
+  [[ "$output" != *"203-qux2"* ]]
+  [[ "$output" != *"204-quux2"* ]]
+}
+
+@test "open+known-error union: mixed fixture enumerates all four (100, 101, 200, 201)" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null'
+  [[ "$output" == *"100-foo.open.md"* ]]
+  [[ "$output" == *"101-bar.known-error.md"* ]]
+  [[ "$output" == *"open/200-foo2.md"* ]]
+  [[ "$output" == *"known-error/201-bar2.md"* ]]
+}
+
+@test "open+known-error union: mixed fixture excludes verifying/parked/closed from BOTH layouts" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null'
+  [[ "$output" != *"102-baz"* ]]
+  [[ "$output" != *"103-qux"* ]]
+  [[ "$output" != *"104-quux"* ]]
+  [[ "$output" != *"202-baz2"* ]]
+  [[ "$output" != *"203-qux2"* ]]
+  [[ "$output" != *"204-quux2"* ]]
+}
+
+# ── Pipeline 3: Verifying-state filter (run-retro Step 4a)
+#
+# SKILL.md prescribes:
+#   ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null
+#
+# run-retro Step 4a uses this to surface verification-close candidates
+# from the session-context evidence. T3 proves the pipeline finds the
+# verifying ticket in BOTH layouts independently.
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "verifying-state pipeline: flat-only fixture finds 102" {
+  build_flat_layout
+  run bash -c 'ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null'
+  [[ "$output" == *"102-baz.verifying.md"* ]]
+}
+
+@test "verifying-state pipeline: per-state-only fixture finds 202" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null'
+  [[ "$output" == *"verifying/202-baz2.md"* ]]
+}
+
+@test "verifying-state pipeline: mixed fixture finds 102 AND 202" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null'
+  [[ "$output" == *"102-baz.verifying.md"* ]]
+  [[ "$output" == *"verifying/202-baz2.md"* ]]
+}
+
+# ── Pipeline 4: ID-anchored ticket lookup
+#   (manage-problem ticket-by-ID, link-incident, close-incident,
+#   transition-problem Step 2, capture-problem Step 2 dup-detect)
+#
+# SKILL.md prescribes:
+#   ls docs/problems/<ID>-*.md docs/problems/*/<ID>-*.md 2>/dev/null
+#
+# T3 exercises lookup of a known-existing ID across both layouts and
+# the missing-ID case (asserts empty stdout + nonzero exit).
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "ID-anchored pipeline: per-state-only fixture finds 200 in subdir" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/200-*.md docs/problems/*/200-*.md 2>/dev/null'
+  [[ "$output" == *"open/200-foo2.md"* ]]
+}
+
+@test "ID-anchored pipeline: mixed fixture finds 100 (flat) and 200 (per-state)" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/100-*.md docs/problems/*/100-*.md 2>/dev/null'
+  [[ "$output" == *"100-foo.open.md"* ]]
+  run bash -c 'ls docs/problems/200-*.md docs/problems/*/200-*.md 2>/dev/null'
+  [[ "$output" == *"open/200-foo2.md"* ]]
+}
+
+@test "ID-anchored pipeline: missing ID yields empty stdout + nonzero exit" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/999-*.md docs/problems/*/999-*.md 2>/dev/null'
+  [ -z "$output" ]
+  [ "$status" -ne 0 ]
+}
+
+# ── Pipeline 5: Brace-expansion ID + state-set (report-upstream)
+#
+# SKILL.md prescribes:
+#   ls docs/problems/<ID>-*.{open,known-error,verifying,closed}.md \
+#      docs/problems/*/<ID>-*.md 2>/dev/null
+#
+# The flat half restricts to a state-set (excludes parked); the
+# per-state half is unrestricted (no state filter). T3 proves the
+# brace expansion composes with the per-state pathspec without false
+# positives across the migration window.
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "brace-id-state pipeline: per-state-only fixture finds 200 via per-state half" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/200-*.{open,known-error,verifying,closed}.md docs/problems/*/200-*.md 2>/dev/null'
+  [[ "$output" == *"open/200-foo2.md"* ]]
+}
+
+@test "brace-id-state pipeline: mixed fixture finds flat 100 (open) and per-state 200 (open)" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/100-*.{open,known-error,verifying,closed}.md docs/problems/*/100-*.md 2>/dev/null'
+  [[ "$output" == *"100-foo.open.md"* ]]
+  run bash -c 'ls docs/problems/200-*.{open,known-error,verifying,closed}.md docs/problems/*/200-*.md 2>/dev/null'
+  [[ "$output" == *"open/200-foo2.md"* ]]
+}
+
+@test "brace-id-state pipeline: parked ticket excluded from flat half but matched in per-state half" {
+  # Flat 103 is parked (excluded by the brace state-set); per-state 203
+  # is parked but the per-state pathspec is state-unrestricted, so it
+  # MUST surface. This is the contract architect finding 2 codifies.
+  build_mixed_layout
+  run bash -c 'ls docs/problems/103-*.{open,known-error,verifying,closed}.md docs/problems/*/103-*.md 2>/dev/null'
+  [[ "$output" != *"103-qux.parked.md"* ]]
+  run bash -c 'ls docs/problems/203-*.{open,known-error,verifying,closed}.md docs/problems/*/203-*.md 2>/dev/null'
+  [[ "$output" == *"parked/203-qux2.md"* ]]
+}
+
+# ── Pipeline 6: Closed-ticket ID-anchored lookup
+#   (review-problems Step 5: closed-section rendering uses
+#   `docs/problems/*.closed.md docs/problems/closed/*.md`)
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "closed-ticket pipeline: mixed fixture enumerates flat 104 and per-state 204" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/*.closed.md docs/problems/closed/*.md 2>/dev/null'
+  [[ "$output" == *"104-quux.closed.md"* ]]
+  [[ "$output" == *"closed/204-quux2.md"* ]]
+}
+
+@test "closed-ticket pipeline: per-state-only fixture excludes other-state subdirs" {
+  build_per_state_layout
+  run bash -c 'ls docs/problems/*.closed.md docs/problems/closed/*.md 2>/dev/null'
+  [[ "$output" == *"closed/204-quux2.md"* ]]
+  [[ "$output" != *"open/200"* ]]
+  [[ "$output" != *"known-error/201"* ]]
+  [[ "$output" != *"verifying/202"* ]]
+  [[ "$output" != *"parked/203"* ]]
+}
+
+# ── Pipeline 7: Parked-state filter (review-problems Step 3 parked section)
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "parked-state pipeline: mixed fixture enumerates flat 103 and per-state 203" {
+  build_mixed_layout
+  run bash -c 'ls docs/problems/*.parked.md docs/problems/parked/*.md 2>/dev/null'
+  [[ "$output" == *"103-qux.parked.md"* ]]
+  [[ "$output" == *"parked/203-qux2.md"* ]]
+}
+
+# ── Composition: ls-with-2>/dev/null exit-code semantics
+# (mirrors T2's empty-fixture contract; proves the SKILL.md contract
+# remains "stdout content is the signal, not exit code")
+# ──────────────────────────────────────────────────────────────────────────────
+
+@test "all pipelines: empty fixture produces empty stdout across every shape" {
+  run bash -c "ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1"
+  [ -z "$output" ]
+  run bash -c 'ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null'
+  [ -z "$output" ]
+  run bash -c 'ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null'
+  [ -z "$output" ]
+  run bash -c 'ls docs/problems/100-*.md docs/problems/*/100-*.md 2>/dev/null'
+  [ -z "$output" ]
+}

package/skills/capture-problem/SKILL.md

@@ -23,7 +23,7 @@ This skill is the foreground-lightweight-capture variant of `/wr-itil:manage-pro
 
 ## Rule 6 audit (per ADR-032 + ADR-013)
 
-This skill has **zero AskUserQuestion branches** by design. Each potentially-interactive decision is framework-mediated per ADR-044:
+This skill has **one classification-only AskUserQuestion (type-tag, taste authority per ADR-044 category 5) and zero control-flow branches keyed on the answer**. Each potentially-interactive decision is framework-mediated per ADR-044:
 
 | Decision | Resolution |
 |----------|-----------|
@@ -32,8 +32,9 @@ This skill has **zero AskUserQuestion branches** by design. Each potentially-int
 | Effort default | Framework-policy: `M` flagged "deferred — re-rate at next /wr-itil:review-problems". |
 | Multi-concern split | Out of scope: capture-problem creates one ticket per invocation. Multi-concern observations route to `/wr-itil:manage-problem` (its Step 4b owns the split). |
 | Empty `$ARGUMENTS` | Halt-with-stderr-directive: print "capture-problem requires a description in $ARGUMENTS — invoke /wr-itil:manage-problem instead for the full intake flow" and exit. AFK orchestrators MUST NOT invoke capture-problem with empty arguments — caller-side contract. |
+| Type classification (P170 / ADR-060 item 8c) | Taste authority per ADR-044 category 5. AskUserQuestion fires for `type` ∈ {`technical`, `user-business`} when no caller-side flag pre-resolved it. `--type=<value>` flag pre-resolves the answer (silent-proceed). `--no-prompt` flag defaults to `technical` (silent-proceed). Maintainer-side ONLY: this prompt is paired with JTBD-301 protection — `.github/ISSUE_TEMPLATE/problem-report.yml` (plugin-user-side intake) MUST NOT carry an equivalent type selector; triage assigns the type during `/wr-itil:manage-problem` ingestion of user-reported issues. **I2 invariant** (ADR-060 line 98): the prompt is a classification facet, not a workflow split — Steps 0-7 control-flow is identical regardless of the chosen `type_value`; only the substituted value in the Step 4 skeleton template differs. |
 
-Per ADR-013 Rule 6 fail-safe: every branch above resolves without user input, so AFK and interactive contexts behave identically.
+Per ADR-013 Rule 6 fail-safe: every decision above resolves without interactive user input in non-interactive contexts (the type-tag carve-out resolves to `technical` via `--no-prompt` or `--type=` caller-side pre-resolution). AFK orchestrators MUST pass `--no-prompt` or `--type=<value>` per JTBD-006 § Persona Constraints; AFK callers that omit both flags violate the caller-side contract. Interactive and pre-resolved AFK paths produce identical observable outputs except for the `**Type**:` field value, satisfying the I2 invariant by construction.
 
 ## Steps
 
@@ -54,18 +55,42 @@ if [ "$reconcile_exit" -eq 1 ]; then
 fi
 ```
 
-### 1. Parse the description from `$ARGUMENTS`
+### 1. Parse the description and flags from `$ARGUMENTS`
 
-The description is the full free-text payload from `$ARGUMENTS`. Empty arguments halts per the Rule 6 audit above.
+`$ARGUMENTS` may carry up to two leading flags before the free-text description (caller-side pre-resolution per ADR-044 silent-proceed shape):
+
+| Flag | Effect on Step 1.5 |
+|------|-------------------|
+| `--type=technical` | Pre-resolves type to `technical`; Step 1.5 skips the AskUserQuestion. |
+| `--type=user-business` | Pre-resolves type to `user-business`; Step 1.5 skips the AskUserQuestion. |
+| `--no-prompt` | Pre-resolves type to `technical` (default); Step 1.5 skips the AskUserQuestion. |
+
+Strip recognised leading flags from `$ARGUMENTS`; the remainder (after flags) is the free-text description. If both `--type=<value>` and `--no-prompt` are present, `--type=<value>` wins (more specific). Unknown leading flags halt-with-stderr-directive: print "capture-problem: unknown flag '<flag>' — recognised flags: --type=technical, --type=user-business, --no-prompt" and exit.
+
+Empty description (post-flag-strip) halts per the Rule 6 audit above.
 
 Derive a kebab-case title slug from the first 8-10 non-stopword tokens of the description (matching the existing `manage-problem` slug derivation pattern).
 
+### 1.5 Type classification (taste authority per ADR-044 category 5)
+
+Resolve `type_value` ∈ {`technical`, `user-business`} per the following framework-mediated dispatch:
+
+1. **If `--type=<value>` was set in Step 1**: use that value; do NOT fire AskUserQuestion (silent-proceed per ADR-013 Rule 5).
+2. **Else if `--no-prompt` was set in Step 1**: default `type_value = technical`; do NOT fire AskUserQuestion. JTBD-006 protection: AFK orchestrators MUST pass this flag (or `--type=<value>`).
+3. **Else** (interactive context, no caller-side pre-resolution): fire AskUserQuestion with options `technical` (default) and `user-business`. Question text: *"What type of problem is this?"* Per-option descriptions:
+   - `technical` — *"Bug, defect, broken behaviour, framework drift — root cause sits in code or process."*
+   - `user-business` — *"Missing capability, UX gap, adopter friction, JTBD-shaped need — root cause sits in unmet user need."*
+
+**I2 invariant guard (ADR-060 line 98)**: the resolved `type_value` is used at Step 4 ONLY as a substituted string in the skeleton template's `**Type**:` body field. Steps 2, 3, 4 (other than the `**Type**:` substitution), 5, 6, 7 execute identically regardless of `type_value`. The skill carries NO control-flow branch keyed on `type` — that would convert classification into a workflow split and violate I2. Pure-bash supporting-script enforcement of this invariant lives in `packages/itil/scripts/test/i2-no-type-branching.bats`; the SKILL.md surface coverage gap is named at P176 (descendant of P012 master harness).
+
+**JTBD-301 scope guard**: this prompt fires on the maintainer-side `/wr-itil:capture-problem` skill only. The plugin-user-side intake (`.github/ISSUE_TEMPLATE/problem-report.yml`) MUST NOT carry an equivalent type selector — plugin-user persona constraint is "no pre-classification". Triage assigns `type` during `/wr-itil:manage-problem` ingestion of user-reported issues, not at user-report time.
+
 ### 2. Minimal-grep duplicate check (3-keyword title-only)
 
-Extract up to **3 distinct kebab-cased non-stopword keywords** from the description. Grep the **filenames** of `docs/problems/*.md` (NOT bodies — title-only is the conservative threshold per architect verdict on Q1):
+Extract up to **3 distinct kebab-cased non-stopword keywords** from the description. Grep the **filenames** of `docs/problems/*.md` AND `docs/problems/<state>/*.md` (NOT bodies — title-only is the conservative threshold per architect verdict on Q1; dual-tolerant per RFC-002 migration window):
 
 ```bash
-match_count=$(ls docs/problems/*.md 2>/dev/null \
+match_count=$(ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null \
               | grep -ciE 'kw1|kw2|kw3' || true)
 ```
 
@@ -90,8 +115,16 @@ The marker is shared between `manage-problem` and `capture-problem` per ADR-032
 Same P056-safe local_max + origin_max formula as `/wr-itil:manage-problem` Step 3:
 
 ```bash
-local_max=$(ls docs/problems/*.md 2>/dev/null | sed 's/.*\///' | grep -oE '^[0-9]+' | sort -n | tail -1)
-origin_max=$(git ls-tree --name-only origin/main docs/problems/ 2>/dev/null | sed 's|^docs/problems/||' | grep -oE '^[0-9]+' | sort -n | tail -1)
+# Dual-tolerant ticket enumeration (RFC-002 migration window). Both
+# halves of the OR contribute to next-ID compute — flat-layout 104 +
+# per-state 204 BOTH appear in `local_max` so the next-ID compute
+# never re-allocates an already-taken ID. Architect finding 2 (RFC-002
+# T2) — capture-problem's next-ID surface is a separate ADR-031
+# contract from generic enumeration; missing the per-state half
+# regresses ID allocation. The `git ls-tree -r` recursive flag
+# extends the same coverage to the origin tree.
+local_max=$(ls docs/problems/*.md docs/problems/*/*.md 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1)
+origin_max=$(git ls-tree -r --name-only origin/main docs/problems/ 2>/dev/null | sed 's|.*/||' | grep -oE '^[0-9]+' | sort -n | tail -1)
 next=$(printf '%03d' $(( $(echo -e "${local_max:-0}\n${origin_max:-0}" | sort -n | tail -1) + 1 )))
 ```
 
@@ -110,10 +143,11 @@ Log the renumber decision in the operation report if origin and local diverged.
 **Reported**: <YYYY-MM-DD>
 **Priority**: 3 (Medium) — Impact: 3 x Likelihood: 1 (deferred — re-rate at next /wr-itil:review-problems)
 **Effort**: M (deferred — re-rate at next /wr-itil:review-problems)
+**Type**: <type_value>
 
 ## Description
 
-<full description from $ARGUMENTS>
+<full description from $ARGUMENTS, with leading recognised flags stripped>
 
 ## Symptoms
 
@@ -192,7 +226,9 @@ The trailing pointer is **not optional** — it is the user-visible signal that
 |---------|----------------|-----------------|
 | Duplicate-check | Wide-net grep + AskUserQuestion branch on matches | 3-keyword title-only grep, list-only (no branch) |
 | Multi-concern split | Step 4b AskUserQuestion | Out of scope (one ticket per invocation) |
-| Skeleton-fill | Full-intake; AskUserQuestion for missing fields | Deferred-placeholder pattern; no AskUserQuestion |
+| Skeleton-fill | Full-intake; AskUserQuestion for missing fields | Deferred-placeholder pattern + one classification-only AskUserQuestion (type-tag) |
+| Type-tag prompt | Step 4-equivalent AskUserQuestion fires alongside other intake fields | Step 1.5 classification-only AskUserQuestion; `--type=` and `--no-prompt` flags pre-resolve for non-interactive callers; I2 invariant: no control-flow branch keyed on type |
+| AskUserQuestion authority | Multiple branches (deviation-approval / direction-setting / taste / mechanical) | Exactly one classification-only fire (taste, ADR-044 cat. 5); zero control-flow branches |
 | README refresh | P094 inline (regenerate + stage in same commit) | Deferred to next `/wr-itil:review-problems` |
 | Status transitions | Step 7 owns Open → Known Error → Verifying → Closed | Out of scope (creation only) |
 | Commit grain | One commit per intake (or per split-concern set) | One commit per capture |
@@ -206,12 +242,17 @@ The two skills share the `/tmp/manage-problem-grep-${SESSION_ID}` create-gate ma
 - **P014** (`docs/problems/014-aside-invocation-for-governance-skills.open.md`) — parent / master tracker.
 - **P078** — capture-on-correction OFFER pattern; depends on capture-problem shipping.
 - **P119** — manage-problem create-gate hook; capture-problem composes with the same marker.
+- **P170** (`docs/problems/170-...open.md`) — RFC framework driver; Slice 4 B7.T3 / item 8c authored the type-classification prompt at Step 1.5.
+- **P176** — agent-side I2 (no type-branching) coverage gap on the SKILL.md surface (this file's surface); descendant of P012 master harness ticket. The Step 1.5 I2 invariant guard is enforced by audit-trailed prose here per ADR-052 § Surface 2 escape-hatch contract; behavioural enforcement awaits the master harness.
 - **ADR-032** (`docs/decisions/032-governance-skill-invocation-patterns.proposed.md`) — foreground-lightweight-capture variant amendment.
 - **ADR-038** — progressive-disclosure pattern (SKILL.md + REFERENCE.md split).
-- **ADR-044** — decision-delegation contract (framework-mediated mechanical-stage carve-outs).
+- **ADR-044** — decision-delegation contract; type classification is taste authority per category 5; `--no-prompt` / `--type=<value>` are policy-authorised silent-proceed shapes per category 4.
 - **ADR-049** — bin/ on PATH; capture-problem reuses the existing `wr-itil-reconcile-readme` shim.
-- **ADR-052** — behavioural-tests-default for skill testing.
+- **ADR-052** — behavioural-tests-default for skill testing; SKILL.md I2 surface coverage gap is named, not silent (P176 + ADR-052 § Surface 2).
+- **ADR-060** (`docs/decisions/060-...accepted.md`) — Phase 1 item 8c authored Step 1.5 here; I2 invariant (line 98) governs the no-control-flow-branch contract; line 132 names the maintainer-side-only / JTBD-301-protection scope; line 160 (Confirmation criterion 4) gates the type-prompt placement.
+- **JTBD-301** (`docs/jtbd/plugin-user/JTBD-301-...md`) — plugin-user no-pre-classification persona constraint; protected by the Step 1.5 maintainer-side scope guard.
 - `packages/itil/skills/manage-problem/SKILL.md` — heavyweight intake counterpart.
 - `packages/itil/skills/review-problems/SKILL.md` — re-rates the deferred placeholders + refreshes README.md.
+- `packages/itil/scripts/test/i2-no-type-branching.bats` — pure-bash supporting-script enforcement of the I2 invariant; this SKILL.md change does not affect any pure-bash script and so does not change the bats outcome (still green).
 
 $ARGUMENTS

package/skills/capture-rfc/SKILL.md

@@ -89,8 +89,10 @@ Derive a kebab-case title slug from the first 8-10 non-stopword tokens of `$desc
 For each `P<NNN>` in the trace list:
 
 ```bash
-# Check existence in any lifecycle status
-trace_files=$(ls docs/problems/<NNN>-*.md 2>/dev/null)
+# Check existence in any lifecycle status (dual-tolerant — RFC-002
+# migration window covers BOTH flat `docs/problems/<NNN>-<title>.<state>.md`
+# AND per-state subdir `docs/problems/<state>/<NNN>-<title>.md` layouts).
+trace_files=$(ls docs/problems/<NNN>-*.md docs/problems/*/<NNN>-*.md 2>/dev/null)
 ```
 
 **I1 hard-block (per ADR-060 § Confirmation criterion 1)**:
@@ -198,7 +200,8 @@ For each problem ID in `$problem_trace`, invoke the helper before commit:
 ```bash
 for pid_token in $(echo "$problem_trace" | tr ',' ' '); do
   pid_num="${pid_token#P}"
-  problem_file=$(ls docs/problems/${pid_num}-*.md 2>/dev/null | head -1)
+  # Dual-tolerant ticket discovery (RFC-002 migration window).
+  problem_file=$(ls docs/problems/${pid_num}-*.md docs/problems/*/${pid_num}-*.md 2>/dev/null | head -1)
   [ -z "$problem_file" ] && continue
   bash "$(wr-itil-script-path 2>/dev/null || echo packages/itil/scripts)/update-problem-rfcs-section.sh" "$problem_file" docs/rfcs
   git add "$problem_file"

package/skills/close-incident/SKILL.md

@@ -63,10 +63,10 @@ Branch on which section is present:
 **Case A — `## Linked Problem` present**:
 
 1. Extract the problem ID `P<NNN>` from the section.
-2. Locate the problem file:
+2. Locate the problem file. Dual-tolerant lookup spans flat layout AND per-state subdir layout (RFC-002 migration window):
 
    ```bash
-   ls docs/problems/<NNN>-*.md 2>/dev/null | head -1
+   ls docs/problems/<NNN>-*.md docs/problems/*/<NNN>-*.md 2>/dev/null | head -1
    ```
 
    Accept both `P<NNN>-*.md` and `<NNN>-*.md` naming conventions for robustness.

package/skills/link-incident/SKILL.md

@@ -42,11 +42,13 @@ The link operation works on any incident status — investigating, mitigating, r
 
 ### 3. Locate the problem file
 
+Dual-tolerant lookup spans flat layout AND per-state subdir layout (RFC-002 migration window):
+
 ```bash
-ls docs/problems/<MMM>-*.md 2>/dev/null | head -1
+ls docs/problems/<MMM>-*.md docs/problems/*/<MMM>-*.md 2>/dev/null | head -1
 ```
 
-Accept any lifecycle suffix: `.open.md`, `.known-error.md`, `.verifying.md`, `.closed.md`.
+Accept any lifecycle suffix: `.open.md`, `.known-error.md`, `.verifying.md`, `.closed.md` in flat layout; bare `<MMM>-<title>.md` under `docs/problems/<state>/` in per-state layout.
 
 - If no file matches, report "No problem `P<MMM>` found. Check `/wr-itil:list-problems` for the current backlog." and exit.
 - Read the problem file's title from the `# Problem <MMM>: <Title>` header line.

package/skills/list-problems/SKILL.md

@@ -12,15 +12,16 @@ This skill is the P071 phased-landing split of `/wr-itil:manage-problem list` pe
 
 ## Scope
 
-Included in the ranking table:
-- `docs/problems/*.open.md` — open tickets (under investigation)
-- `docs/problems/*.known-error.md` — known errors (root cause confirmed, fix NOT yet released)
+Included in the ranking table (RFC-002 migration window — each glob is dual-tolerant, covering BOTH the flat `docs/problems/<NNN>-<title>.<state>.md` filename-suffix layout AND the per-state subdir `docs/problems/<state>/<NNN>-<title>.md` layout):
+
+- `docs/problems/*.open.md` + `docs/problems/open/*.md` — open tickets (under investigation)
+- `docs/problems/*.known-error.md` + `docs/problems/known-error/*.md` — known errors (root cause confirmed, fix NOT yet released)
 
 Shown in separate sections, excluded from the dev-work WSJF ranking per ADR-022:
-- `docs/problems/*.verifying.md` — Verification Pending (fix released, awaiting user verification; WSJF multiplier 0)
-- `docs/problems/*.parked.md` — Parked on upstream or user-suspended (WSJF multiplier 0)
+- `docs/problems/*.verifying.md` + `docs/problems/verifying/*.md` — Verification Pending (fix released, awaiting user verification; WSJF multiplier 0)
+- `docs/problems/*.parked.md` + `docs/problems/parked/*.md` — Parked on upstream or user-suspended (WSJF multiplier 0)
 
-`docs/problems/*.closed.md` is omitted entirely (the view is of active backlog, not the closed archive).
+`docs/problems/*.closed.md` + `docs/problems/closed/*.md` is omitted entirely (the view is of active backlog, not the closed archive).
 
 ## Steps
 
@@ -31,7 +32,10 @@ Reuse the same `git log`-based freshness test as `/wr-itil:manage-problem review
 ```bash
 readme_commit=$(git log -1 --format=%H -- docs/problems/README.md 2>/dev/null)
 if [ -z "$readme_commit" ] || \
-   git log --oneline "${readme_commit}..HEAD" -- 'docs/problems/*.md' ':!docs/problems/README.md' 2>/dev/null | grep -q .; then
+   git log --oneline "${readme_commit}..HEAD" -- 'docs/problems/*.md' 'docs/problems/*/*.md' ':!docs/problems/README.md' 2>/dev/null | grep -q .; then
+  # Pathspec pair `'docs/problems/*.md' 'docs/problems/*/*.md'` is the
+  # RFC-002 dual-tolerant transitional shape — covers BOTH the flat
+  # layout AND the per-state subdir layout.
   echo "stale"
 fi
 ```
@@ -42,12 +46,12 @@ fi
 
 ### 2. Live scan (cache-stale fallback)
 
-Enumerate the backlog files via glob:
+Enumerate the backlog files via dual-tolerant globs (RFC-002 migration window — each line covers BOTH the flat `<NNN>-<title>.<state>.md` filename-suffix layout AND the per-state subdir `<state>/<NNN>-<title>.md` layout):
 
 ```bash
-ls docs/problems/*.open.md docs/problems/*.known-error.md 2>/dev/null
-ls docs/problems/*.verifying.md 2>/dev/null  # for Verification Queue section
-ls docs/problems/*.parked.md 2>/dev/null     # for Parked section
+ls docs/problems/*.open.md docs/problems/*.known-error.md docs/problems/open/*.md docs/problems/known-error/*.md 2>/dev/null
+ls docs/problems/*.verifying.md docs/problems/verifying/*.md 2>/dev/null  # for Verification Queue section
+ls docs/problems/*.parked.md docs/problems/parked/*.md 2>/dev/null        # for Parked section
 ```
 
 For each `.open.md` and `.known-error.md` file, read the `**Status**`, `**Priority**`, `**Effort**`, and `**WSJF**` lines from the frontmatter section. Compute WSJF if missing: `WSJF = (Severity × StatusMultiplier) / EffortDivisor` per `/wr-itil:manage-problem` WSJF Prioritisation. Default to M (divisor 2) when Effort is absent; flag missing scores so the user knows a review is overdue.