@mmerterden/multi-agent-pipeline 10.7.3 → 10.7.4

package/pipeline/skills/shared/core/multi-agent-finish/SKILL.md

@@ -13,7 +13,7 @@ You already wrote (and maybe hand-tested) the change on the current branch  -  o
 
 ```
 Phase 0: Init          → project/branch detect, resolve base + diff (work already done), Jira id, state (NO worktree)
-Phase 4: Review        → deterministic gates + parallel review + Opus triage
+Phase 4: Review        → deterministic gates + parallel review + Fable triage
 Phase 5: Build+Test    → stack-aware build + run existing tests; SUCCESS required (automated gate, not the interactive user-test)
 Phase 6: Commit        → commit remaining changes + push + open PR if none exists
 Phase 7: Report        → technical analysis + Jira comment with test scenarios (channels: Jira / PR / Confluence / Wiki)

package/pipeline/skills/shared/core/multi-agent-help/SKILL.md

@@ -59,11 +59,11 @@ How It Works (Phase 0  -  Interactive Flow):
 Pipeline (after Phase 0):
 
   Phase 0: Init      -> The 8 steps above
-  Phase 1: Analysis  -> Stack detection + codebase scan (Opus)
+  Phase 1: Analysis  -> Stack detection + codebase scan (Fable)
   Phase 2: Planning  -> Task breakdown + architecture review + Plan Approval Gate
   Phase 3: Dev       -> TDD: test -> code -> build (Sonnet) + build queue
-  Phase 4: Review    -> Deterministic gates + parallel AI review + Opus triage
-                        (Claude Code: Opus + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
+  Phase 4: Review    -> Deterministic gates + parallel AI review + Fable triage
+                        (Claude Code: Fable + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
   Phase 5: Test      -> Optional: switch to branch, test in Xcode
   Phase 6: Commit    -> Commit -> push -> PR + issue body update (never auto-closes)
   Phase 7: Report    -> Channels dispatcher (PR · Jira · Confluence · Wiki, multi-select)
@@ -75,7 +75,7 @@ Pipeline (after Phase 0):
 
 Modes:
 
-  (normal)             Full 8 phases, parallel review + Opus triage
+  (normal)             Full 8 phases, parallel review + Fable triage
   --dev                Fast: Init -> Dev(Opus) -> Commit -> Report (no plan gate)
   --local              No worktree  -  works directly on local branch
   autopilot            Skip all confirmations (EXCEPT Phase 7 channels menu)
@@ -194,11 +194,11 @@ Nasıl Çalışır (Phase 0  -  İnteraktif Akış):
 Pipeline (Phase 0'dan sonra):
 
   Phase 0: Init      -> Yukarıdaki 8 adım
-  Phase 1: Analysis  -> Stack tespiti + codebase taraması (Opus)
+  Phase 1: Analysis  -> Stack tespiti + codebase taraması (Fable)
   Phase 2: Planning  -> Task kırılımı + mimari inceleme + Plan Onay Kapısı
   Phase 3: Dev       -> TDD: test -> kod -> build (Sonnet) + build queue
-  Phase 4: Review    -> Deterministik kapılar + paralel AI review + Opus triage
-                        (Claude Code: Opus + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
+  Phase 4: Review    -> Deterministik kapılar + paralel AI review + Fable triage
+                        (Claude Code: Fable + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
   Phase 5: Test      -> Opsiyonel: branch'e geç, Xcode'da test
   Phase 6: Commit    -> Commit -> push -> PR + issue body güncelleme (hiç auto-close yok)
   Phase 7: Report    -> Channels dispatcher (PR · Jira · Confluence · Wiki, multi-select)
@@ -210,7 +210,7 @@ Pipeline (Phase 0'dan sonra):
 
 Modlar:
 
-  (normal)             Tam 8 faz, paralel review + Opus triage
+  (normal)             Tam 8 faz, paralel review + Fable triage
   --dev                Hızlı: Init -> Dev(Opus) -> Commit -> Report (plan gate yok)
   --local              Worktree yok  -  doğrudan local branch'te çalışır
   autopilot            Tüm onayları atla (İSTİSNA: Phase 7 channels menüsü)

package/pipeline/skills/shared/core/multi-agent-review/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: multi-agent-review
 language: en
-description: "Run parallel review on the current branch's diff: 2 models on Claude Code (Opus + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). Review-only slice of the pipeline."
+description: "Run parallel review on the current branch's diff: 2 models on Claude Code (Fable + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). Review-only slice of the pipeline."
 user-invocable: true
 argument-hint: "[branch]  -  optional: branch to review. If omitted, the current branch is used."
 ---
@@ -27,13 +27,13 @@ Skip Phase 0-3 and review the current diff only.
 3. **Start parallel reviewers**  -  reviewer set depends on the host CLI:
 
    **Claude Code (2 in parallel):**
-   - Agent 1: `claude-opus-4.6` → security + architecture
-   - Agent 2: `claude-sonnet-4.6` → general quality
+   - Agent 1: `claude-fable-5` → security + architecture
+   - Agent 2: `claude-sonnet-4-6` → general quality
 
    **Copilot CLI (3 in parallel):**
-   - Agent 1: `claude-opus-4.6` → security + architecture
+   - Agent 1: `claude-opus-4-8` → security + architecture (Fable 5 is not offered on Copilot CLI)
    - Agent 2: `gpt-5.4` → edge cases, different perspective
-   - Agent 3: `claude-sonnet-4.6` → general quality
+   - Agent 3: `claude-sonnet-4-6` → general quality
 
 4. **Store-compliance cross-reference**  -  if iOS/Android release-relevant files changed, the matching catalog is loaded:
 

package/pipeline/skills/shared/core/multi-agent-sync/SKILL.md

@@ -33,7 +33,7 @@ Run all steps automatically:
 Step 0: FIGMA_SYNC  (opt-in) pipeline/scripts/sync-figma-source.sh
                     -- incremental pull from upstream if figmaSource.path is set
 Step 1: DETECT      Compare timestamps, find stale targets
-Step 2: COPILOT     Claude Code -> Copilot CLI (instructions + 26 sub-command skills)
+Step 2: COPILOT     Claude Code -> Copilot CLI (instructions + 35 sub-command skills)
 Step 3: REPO        Claude Code -> pipeline repo (genericized, personal data scrub)
 Step 4: WEBSITE     Version + phase/model counts -> {website-host} (i18n + projects.ts)
 Step 5: REMOTE      Pipeline references -> remote-control README
@@ -158,12 +158,14 @@ When invoked with the `release` argument:
 |-------------|-------------|
 | `~/.claude/commands/multi-agent/{cmd}.md` | `~/.copilot/skills/multi-agent-{cmd}/SKILL.md` |
 
-**26 commands are synced** (v7.0.0+ canonical inventory  -  must match `cross-cli-contract.md` section1, drift = contract violation):
+**35 commands are synced** (canonical inventory  -  must match `cross-cli-contract.md` section 1; drift = contract violation):
 
 ```
-autopilot, channels, dev, dev-autopilot, dev-local, dev-local-autopilot,
-help, issue, jira, kill, local, local-autopilot, log, manual-test, purge,
-refactor, resume, review, scan, search, setup, stack, status, sync, test, update
+analysis, analysis-resolve, autopilot, build-optimize, channels, delete, dev,
+dev-autopilot, dev-local, dev-local-autopilot, diff-explain, finish, garbage-collect,
+help, issue, jira, kill, language, local, local-autopilot, log, manual-test,
+prune-logs, purge, refactor, resume, review, scan, search, setup, stack, status,
+sync, test, update
 ```
 
 **NOT synced**: `refs/*`  -  Lazy-load references, Claude Code specific

package/pipeline/scripts/smoke-readme-counts.sh

@@ -1,120 +0,0 @@
-#!/usr/bin/env bash
-# smoke-readme-counts.sh  -  fail when README "Current at-a-glance" counts drift
-# from the actual filesystem. The table is hand-edited; without this gate it
-# silently goes stale every time a smoke / skill / schema is added or removed.
-#
-# Each row is checked independently. Compute the live value, parse the README
-# value, fail if they differ. Output the exact table line so the fix is a
-# one-line edit.
-#
-# Exit 0 = every count matches reality, 1 = at least one row drifted.
-
-set -uo pipefail
-
-ROOT="$(cd "$(dirname "$0")/../.." && pwd)"
-README="$ROOT/README.md"
-
-pass=0
-fail=0
-failures=()
-record_pass() { pass=$((pass + 1)); printf '  \033[0;32mPASS\033[0m %s\n' "$1"; }
-record_fail() { fail=$((fail + 1)); failures+=("$1"); printf '  \033[0;31mFAIL\033[0m %s\n' "$1"; }
-
-printf '→ smoke-readme-counts: README "at-a-glance" table matches filesystem\n'
-
-[ -f "$README" ] || { record_fail "README.md missing"; exit 1; }
-
-# Helper: extract the integer at the end of a markdown table row whose first
-# cell starts with the given prefix. Examples of expected rows:
-#   "| Smoke suites | 73 |"
-#   "| Total `SKILL.md` files across all groups | 195 |"
-parse_count() {
-  local prefix_pattern="$1"
-  # Use grep -E for the line match (more predictable than awk's match()),
-  # then strip everything that isn't the trailing integer cell.
-  local row
-  row=$(grep -E "$prefix_pattern" "$README" | head -1)
-  [ -z "$row" ] && { echo "PARSE_ERROR"; return; }
-  # The integer is the last `| <digits> |` cell on the row.
-  echo "$row" | grep -oE '\| *[0-9]+ *\|' | tail -1 | grep -oE '[0-9]+'
-}
-
-check_count() {
-  local label="$1" actual="$2" prefix_pattern="$3"
-  local declared
-  declared=$(parse_count "$prefix_pattern")
-  if [ -z "$declared" ] || [ "$declared" = "PARSE_ERROR" ]; then
-    record_fail "$label: README row missing or unparseable (pattern: $prefix_pattern)"
-    return
-  fi
-  if [ "$declared" = "$actual" ]; then
-    record_pass "$label: README $declared matches actual $actual"
-  else
-    record_fail "$label: README claims $declared but actual is $actual"
-  fi
-}
-
-# --- Compute live counts -----------------------------------------------------
-
-# Slash commands: every .md in commands/multi-agent/ that is NOT an internal
-# `_`-prefixed picker fragment.
-SLASH=$(ls "$ROOT"/pipeline/commands/multi-agent/*.md 2>/dev/null \
-  | xargs -n1 basename | grep -v '^_' | wc -l | tr -d ' ')
-
-# Copilot skills: skill dirs whose SKILL.md frontmatter is `name: multi-agent-*`
-# AND `user-invocable: true`. Mirrors the Claude Code slash-command surface.
-COPILOT=$(grep -lE '^name: multi-agent-' "$ROOT"/pipeline/skills/shared/core/*/SKILL.md 2>/dev/null \
-  | xargs grep -l '^user-invocable: true' 2>/dev/null | wc -l | tr -d ' ')
-
-# Figma skills: SKILL.md count under figma-{common,ios,android}.
-FIGMA=$(find "$ROOT"/pipeline/skills/figma-common "$ROOT"/pipeline/skills/figma-ios "$ROOT"/pipeline/skills/figma-android \
-  -name SKILL.md 2>/dev/null | wc -l | tr -d ' ')
-
-# External skill catalog: subdirectories of shared/external/.
-EXTERNAL=$(ls -d "$ROOT"/pipeline/skills/shared/external/*/ 2>/dev/null | wc -l | tr -d ' ')
-
-# Total SKILL.md files: every SKILL.md anywhere under pipeline/skills/.
-TOTAL_SKILLS=$(find "$ROOT"/pipeline/skills -name SKILL.md 2>/dev/null | wc -l | tr -d ' ')
-
-# Smoke suites: every smoke-*.sh in pipeline/scripts/.
-SMOKES=$(ls "$ROOT"/pipeline/scripts/smoke-*.sh 2>/dev/null | wc -l | tr -d ' ')
-
-# Eval-triage fixtures: subdirs in pipeline/eval/triage/ whose names start with
-# a digit (the numbered fixture pattern, e.g. `01-empty-findings`).
-EVAL_TRIAGE=$(ls -d "$ROOT"/pipeline/eval/triage/[0-9]* 2>/dev/null | wc -l | tr -d ' ')
-
-# JSON schemas (excluding token-budget.json which is a config, not a schema).
-SCHEMAS=$(ls "$ROOT"/pipeline/schemas/*.schema.json 2>/dev/null | wc -l | tr -d ' ')
-
-# Agent personas: every .md under pipeline/agents/.
-AGENTS=$(find "$ROOT"/pipeline/agents -name '*.md' 2>/dev/null | wc -l | tr -d ' ')
-
-# Store-compliance skills (apple-archive-compliance + google-play-compliance).
-COMPLIANCE=$(ls -d "$ROOT"/pipeline/skills/shared/core/apple-archive-compliance "$ROOT"/pipeline/skills/shared/core/google-play-compliance 2>/dev/null | wc -l | tr -d ' ')
-
-# Golden-task fixtures.
-GOLDEN=$(ls -d "$ROOT"/pipeline/eval/golden-tasks/[0-9]* 2>/dev/null | wc -l | tr -d ' ')
-
-# --- Verify each row ---------------------------------------------------------
-
-check_count "slash commands"        "$SLASH"        '^\| Slash commands \(colon-form'
-check_count "Copilot skills"        "$COPILOT"      '^\| Copilot skills \(dash-form'
-check_count "store-compliance"      "$COMPLIANCE"   '^\| Store-compliance skills'
-check_count "figma skills"          "$FIGMA"        '^\| Figma skills'
-check_count "external catalog"      "$EXTERNAL"     '^\| External skill catalog'
-check_count "total SKILL.md"        "$TOTAL_SKILLS" '^\| Total `SKILL\.md`'
-check_count "smoke suites"          "$SMOKES"       '^\| Smoke suites'
-check_count "golden-task fixtures"  "$GOLDEN"       '^\| Golden-task fixtures'
-check_count "eval-triage fixtures"  "$EVAL_TRIAGE"  '^\| Eval-triage fixtures'
-check_count "JSON schemas"          "$SCHEMAS"      '^\| JSON schemas'
-check_count "agent personas"        "$AGENTS"       '^\| Agent personas'
-
-# --- Summary -----------------------------------------------------------------
-total=$((pass + fail))
-printf '\n→ smoke-readme-counts: %d/%d passed\n' "$pass" "$total"
-if [ "$fail" -ne 0 ]; then
-  printf '\nFailures (one-line README fix per row):\n'
-  for f in "${failures[@]}"; do printf '  - %s\n' "$f"; done
-  exit 1
-fi
-exit 0