@ai-dev-methodologies/rlp-desk 0.5.0 → 0.5.1

package/docs/plans/keen-sauteeing-snowflake.md

@@ -0,0 +1,245 @@
+# CB 정합성 수정 + 분석 로그 항상 생성
+
+## Context
+
+1. CB_THRESHOLD=3에서 모델 업그레이드 경로(3~4단계) 작동 불가 — 설계 결함
+2. campaign.jsonl, metadata.json이 `--debug` 전용 — 기본 실행에서 분석 데이터 없음
+3. campaign.jsonl에 분석에 필요한 필드 부족 (consecutive_failures, model_upgraded 등)
+
+---
+
+## 변경 1: CB_THRESHOLD 기본값 3 → 6
+
+**파일**: `src/scripts/run_ralph_desk.zsh:75`
+
+```diff
+- CB_THRESHOLD="${CB_THRESHOLD:-3}"
++ CB_THRESHOLD="${CB_THRESHOLD:-6}"
+```
+
+**파일**: `src/governance.md` §8 — CB 기본값 6 반영
+
+---
+
+## 변경 2: campaign.jsonl, metadata.json 항상 생성 (debug 게이팅 제거)
+
+### 2a. analytics 디렉토리 항상 생성
+
+**파일**: `src/scripts/run_ralph_desk.zsh` (~L1890)
+
+```diff
+- # --- Analytics directory: create only when --debug or --with-self-verification ---
+- if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
+-   mkdir -p "$ANALYTICS_DIR" 2>/dev/null
+- fi
++ # --- Analytics directory: always create (campaign.jsonl + metadata.json are always-on) ---
++ mkdir -p "$ANALYTICS_DIR" 2>/dev/null
+```
+
+### 2b. metadata.json 항상 작성
+
+**파일**: `src/scripts/run_ralph_desk.zsh` (~L1915-1940)
+
+```diff
+- # --- metadata.json: write at campaign start ---
+- if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
+-   jq -n \
+-     ...
+- fi
++ # --- metadata.json: always write at campaign start (cross-project identification) ---
++ jq -n \
++   ...
++   --arg project_name "$(basename "$ROOT")" \
++   ...
+```
+
+metadata.json에 `project_name` 필드 추가 (basename of ROOT).
+
+### 2c. write_campaign_jsonl() debug 게이팅 제거
+
+**파일**: `src/scripts/lib_ralph_desk.zsh:356`
+
+```diff
+- write_campaign_jsonl() {
+-   if (( ! DEBUG )) && (( ! WITH_SELF_VERIFICATION )); then return 0; fi
++ write_campaign_jsonl() {
+```
+
+### 2d. campaign.jsonl 레코드에 분석 필드 추가
+
+**파일**: `src/scripts/lib_ralph_desk.zsh` write_campaign_jsonl()
+
+추가 필드:
+- `consecutive_failures`: 현재 연속 실패 카운트
+- `model_upgraded`: 이 iteration에서 모델 업그레이드 발생 여부 (0/1)
+- `fix_contract`: 이전 iteration의 fix contract 존재 여부 (0/1)
+
+```diff
+  '{iter: $iter, us_id: $us_id, worker_model: $worker_model, ...}'
++ 에 --argjson consecutive_failures "$CONSECUTIVE_FAILURES"
++    --argjson model_upgraded "$_MODEL_UPGRADED"
+```
+
+### 2e. campaign.jsonl 버전 관리도 항상 적용
+
+**파일**: `src/scripts/run_ralph_desk.zsh` (~L1904-1913)
+
+```diff
+- # --- campaign.jsonl versioning (in analytics dir, after mkdir) ---
+- if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
+-   if [[ -f "$CAMPAIGN_JSONL" ]]; then
++ # --- campaign.jsonl versioning (always-on) ---
++ if [[ -f "$CAMPAIGN_JSONL" ]]; then
+```
+
+debug.log 버전 관리는 `--debug` 게이팅 유지 (debug.log 자체가 debug 전용이므로).
+
+---
+
+## 변경 3: `--with-self-verification` SV report 경로 버그 수정
+
+**현재 버그:** SV report 경로 불일치
+- **쓰기** (`lib_ralph_desk.zsh:553-556`): `$LOGS_DIR/self-verification-report-NNN.md` (프로젝트 로컬)
+- **읽기** (`lib_ralph_desk.zsh:434`): `$ANALYTICS_DIR/self-verification-report-*.md` (홈)
+- 쓰는 곳과 읽는 곳이 다르므로 campaign-report.md에서 SV report 참조 실패
+
+**수정:** 읽기 경로를 `$LOGS_DIR`로 통일 (프로젝트 로컬이 맞음 — iteration 아티팩트를 분석한 결과물)
+
+**파일**: `src/scripts/lib_ralph_desk.zsh:434`
+
+```diff
+-     sv_report=$(ls -t "$ANALYTICS_DIR"/self-verification-report-*.md 2>/dev/null | head -1)
++     sv_report=$(ls -t "$LOGS_DIR"/self-verification-report-*.md 2>/dev/null | head -1)
+```
+
+**governance §6 문서에서도 정리:**
+- `self-verification-report-NNN.md` → 프로젝트 로컬 `logs/<slug>/` 에 명시
+- `self-verification-data.json` → 코드에서 생성 안 함 (governance §6에만 명시, agent-mode 전용). 문서에서 "(agent-mode only)" 주석 추가
+
+### `--with-self-verification` 파일 위치 정리
+
+| 파일 | 위치 | 생성 조건 | 용도 |
+|------|------|-----------|------|
+| `self-verification-report-NNN.md` | 프로젝트 로컬 `logs/<slug>/` | `--with-self-verification` | claude CLI로 iteration 아티팩트 분석한 서술형 리포트 |
+| `self-verification-data.json` | 홈 `analytics/<slug>--<hash>/` | agent-mode + `--with-self-verification` | 구조화된 SV 데이터 (tmux 모드에서는 생성 안 됨) |
+
+---
+
+## 변경하지 않는 것
+
+| 항목 | 위치 | 이유 |
+|------|------|------|
+| `iter-NNN.*` 전체 | 프로젝트 로컬 | 프로젝트 코드/git과 직접 연결 |
+| `campaign-report.md` | 프로젝트 로컬 | git diff stat 참조 |
+| `cost-log.jsonl` | 프로젝트 로컬 | campaign-report가 참조 |
+| `runtime/` | 프로젝트 로컬 | 실시간 운영 데이터 |
+| `baseline.log` | 프로젝트 로컬 | 프로젝트 기준점 |
+| `SV report` | 프로젝트 로컬 | iteration 아티팩트 분석 결과물 |
+| `debug.log` | 홈, `--debug` 전용 | verbose, 필요할 때만 |
+
+---
+
+## 수정 대상 파일
+
+| 파일 | 변경 |
+|------|------|
+| `src/scripts/run_ralph_desk.zsh` | CB default 6, analytics dir 항상 생성, metadata.json 항상 쓰기 + project_name, campaign.jsonl 버전 관리 항상 적용 |
+| `src/scripts/lib_ralph_desk.zsh` | write_campaign_jsonl() debug 게이팅 제거 + 필드 추가, SV report 읽기 경로 수정 |
+| `src/governance.md` | §8 CB 기본값 6, §6 파일 구조 + SV 위치 정리 |
+
+---
+
+## Verification (TDD)
+
+모든 변경은 **테스트 먼저 작성 → RED 확인 → 구현 → GREEN 확인** 순서로 진행.
+
+테스트 파일: `tests/test_cb_and_analytics.sh`
+
+### 테스트 목록 (RED → GREEN 순서대로)
+
+**변경 1: CB_THRESHOLD**
+```bash
+# T1: CB 기본값이 6인지 확인
+test_cb_default_is_6() {
+  source src/scripts/run_ralph_desk.zsh --dry-run 2>/dev/null  # 또는 grep
+  assert CB_THRESHOLD == 6
+}
+
+# T2: consensus 모드에서 effective CB가 12(6*2)인지 확인
+test_cb_consensus_doubles_to_12() {
+  VERIFY_CONSENSUS=1 source ...
+  assert EFFECTIVE_CB_THRESHOLD == 12
+}
+```
+
+**변경 2: campaign.jsonl 항상 생성**
+```bash
+# T3: --debug 없이 analytics 디렉토리 생성 확인
+test_analytics_dir_created_without_debug() {
+  DEBUG=0 WITH_SELF_VERIFICATION=0
+  # run init section → assert mkdir -p "$ANALYTICS_DIR" called
+  assert -d "$ANALYTICS_DIR"
+}
+
+# T4: --debug 없이 metadata.json 생성 확인
+test_metadata_written_without_debug() {
+  DEBUG=0 WITH_SELF_VERIFICATION=0
+  # run metadata write section
+  assert -f "$METADATA_FILE"
+}
+
+# T5: metadata.json에 project_name 필드 존재
+test_metadata_has_project_name() {
+  assert jq -r '.project_name' "$METADATA_FILE" != "null"
+}
+
+# T6: write_campaign_jsonl()이 --debug 없이 쓰는지 확인
+test_campaign_jsonl_written_without_debug() {
+  DEBUG=0 WITH_SELF_VERIFICATION=0
+  write_campaign_jsonl 1 "US-001" "pass"
+  assert -f "$CAMPAIGN_JSONL"
+}
+
+# T7: campaign.jsonl 레코드에 consecutive_failures 필드 존재
+test_campaign_jsonl_has_consecutive_failures() {
+  CONSECUTIVE_FAILURES=2
+  write_campaign_jsonl 1 "US-001" "fail"
+  assert jq -r '.consecutive_failures' last_line == 2
+}
+
+# T8: campaign.jsonl 레코드에 model_upgraded 필드 존재
+test_campaign_jsonl_has_model_upgraded() {
+  _MODEL_UPGRADED=1
+  write_campaign_jsonl 1 "US-001" "fail"
+  assert jq -r '.model_upgraded' last_line == 1
+}
+
+# T9: campaign.jsonl 재실행 시 버전 관리 (--debug 없이)
+test_campaign_jsonl_versioned_without_debug() {
+  echo '{}' > "$CAMPAIGN_JSONL"
+  # run versioning section
+  assert -f "${CAMPAIGN_JSONL%.jsonl}-v1.jsonl"
+}
+```
+
+**변경 3: SV report 경로 수정**
+```bash
+# T10: generate_campaign_report()가 $LOGS_DIR에서 SV report 찾는지 확인
+test_sv_report_read_from_logs_dir() {
+  WITH_SELF_VERIFICATION=1
+  touch "$LOGS_DIR/self-verification-report-001.md"
+  # generate_campaign_report 호출
+  # campaign-report.md 안에 $LOGS_DIR 경로의 SV report 참조 확인
+  assert grep "self-verification-report" "$LOGS_DIR/campaign-report.md"
+}
+```
+
+### 실행 순서
+
+1. 테스트 파일 작성 (`tests/test_cb_and_analytics.sh`)
+2. 전체 RED 확인 (모든 테스트 fail)
+3. 변경 1 구현 → T1, T2 GREEN
+4. 변경 2 구현 → T3~T9 GREEN
+5. 변경 3 구현 → T10 GREEN
+6. governance 문서 업데이트
+7. 기존 테스트 회귀 확인: `bash tests/test_us005_tmux_docs.sh`

package/docs/plans/toasty-whistling-diffie.md

@@ -1,4 +1,4 @@
-# Remaining Work Plan — rlp-desk Post v0.5
+# Hotfix: Revert keybinding changes from init_ralph_desk.zsh
 
 **Created**: 2026-03-30
 **Updated**: 2026-03-30 (Phase 1-4 구현 완료, 커밋 대기)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-dev-methodologies/rlp-desk",
-  "version": "0.5.0",
+  "version": "0.5.1",
   "description": "Fresh-context iterative loops for Claude Code — autonomous task completion with independent verification",
   "scripts": {
     "postinstall": "node scripts/postinstall.js",

package/src/governance.md

@@ -185,7 +185,7 @@ Verification occurs at two boundaries, not as a single final event.
 ### Checkpoint 2: Release Readiness (us_id=ALL)
 - Trigger: all individual US pass Checkpoint 1 → Worker signals verify with us_id = "ALL"
 - Scope: all AC + L2 integration (if applicable) + L3 E2E Simulation + L4 deploy (if applicable) + mutation score (if CRITICAL, when mutation testing tool is configured in test-spec)
-- On fail: fix loop; escalation to user if 3 consecutive failures
+- On fail: fix loop; escalation to user if 6 consecutive failures (default cb_threshold)
 
 ### Relationship to Existing Flow
 - Checkpoint 1 = existing per-US verify (§7a). No change.
@@ -574,7 +574,7 @@ In tmux mode: Leader writes `<slug>-escalation.md` with the report and sets BLOC
 |-----------|---------|
 | context-latest.md unchanged for 3 consecutive iterations | BLOCKED |
 | Same acceptance criterion fails 2 consecutive iterations | Upgrade model, retry once (Agent mode only; tmux: same model retry); if still failing → Architecture Escalation (§7¾) → BLOCKED |
-| `cb_threshold` consecutive **fail** verdicts on `cb_threshold` unique criterion IDs | Upgrade to opus, retry once; if still failing → BLOCKED (adjustable via `--cb-threshold`) |
+| `cb_threshold` (default: 6) consecutive **fail** verdicts on `cb_threshold` unique criterion IDs | Upgrade to opus, retry once; if still failing → BLOCKED (adjustable via `--cb-threshold`) |
 | max_iter reached | TIMEOUT (report to user) |
 
 The Leader tracks `consecutive_failures` in `status.json`:

package/src/scripts/lib_ralph_desk.zsh

@@ -351,9 +351,8 @@ write_cost_log() {
   echo '{"iteration":'"$iter"',"estimated_tokens":'"$estimated_tokens"',"token_source":"estimated","prompt_bytes":'"$prompt_bytes"',"claim_bytes":'"$claim_bytes"',"verdict_bytes":'"$verdict_bytes"',"worker_start_time":"'"$worker_start_time"'","worker_end_time":"'"$worker_end_time"'","worker_duration_s":'"$worker_duration_s"',"verifier_start_time":"'"$verifier_start_time"'","verifier_end_time":"'"$verifier_end_time"'","verifier_duration_s":'"$verifier_duration_s"''"$consensus_fields"',"timestamp":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'"}' >> "$COST_LOG"
 }
 
-# --- Analytics: write per-iteration structured data to campaign.jsonl ---
+# --- Analytics: write per-iteration structured data to campaign.jsonl (always-on) ---
 write_campaign_jsonl() {
-  if (( ! DEBUG )) && (( ! WITH_SELF_VERIFICATION )); then return 0; fi
   local iter="$1"
   local us_id="${2:-unknown}"
   local verdict="${3:-unknown}"
@@ -376,12 +375,14 @@ write_campaign_jsonl() {
     --arg claude_verdict "${CLAUDE_VERDICT:-$verdict}" \
     --arg codex_verdict "${CODEX_VERDICT:-N/A}" \
     --argjson consensus "$VERIFY_CONSENSUS" \
+    --argjson consecutive_failures "$CONSECUTIVE_FAILURES" \
+    --argjson model_upgraded "${_MODEL_UPGRADED:-0}" \
     --argjson duration_worker_s "$worker_duration_s" \
     --argjson duration_verifier_s "$verifier_duration_s" \
     --arg project_root "$ROOT" \
     --arg slug "$SLUG" \
     --arg timestamp "$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
-    '{iter: $iter, us_id: $us_id, worker_model: $worker_model, worker_engine: $worker_engine, verifier_engine: $verifier_engine, claude_verdict: $claude_verdict, codex_verdict: $codex_verdict, consensus: $consensus, duration_worker_s: $duration_worker_s, duration_verifier_s: $duration_verifier_s, project_root: $project_root, slug: $slug, timestamp: $timestamp}' \
+    '{iter: $iter, us_id: $us_id, worker_model: $worker_model, worker_engine: $worker_engine, verifier_engine: $verifier_engine, claude_verdict: $claude_verdict, codex_verdict: $codex_verdict, consensus: $consensus, consecutive_failures: $consecutive_failures, model_upgraded: $model_upgraded, duration_worker_s: $duration_worker_s, duration_verifier_s: $duration_verifier_s, project_root: $project_root, slug: $slug, timestamp: $timestamp}' \
     >> "$CAMPAIGN_JSONL"
 }
 
@@ -431,7 +432,7 @@ ${untracked}"
   local sv_summary=""
   if (( WITH_SELF_VERIFICATION )); then
     local sv_report
-    sv_report=$(ls -t "$ANALYTICS_DIR"/self-verification-report-*.md 2>/dev/null | head -1)
+    sv_report=$(ls -t "$LOGS_DIR"/self-verification-report-*.md 2>/dev/null | head -1)
     if [[ -n "$sv_report" ]]; then
       sv_summary="See: $sv_report"
     else

package/src/scripts/run_ralph_desk.zsh

@@ -72,7 +72,7 @@ VERIFY_CONSENSUS="${VERIFY_CONSENSUS:-0}"   # 0|1
 FINAL_CONSENSUS="${FINAL_CONSENSUS:-0}"     # 0|1 — consensus for final ALL verify only (independent of VERIFY_CONSENSUS)
 CONSENSUS_SCOPE="${CONSENSUS_SCOPE:-all}"   # all|final-only
 CONSENSUS_FAIL_FAST="${CONSENSUS_FAIL_FAST:-0}" # 0|1 — skip second verifier if first fails
-CB_THRESHOLD="${CB_THRESHOLD:-3}"           # consecutive failures before BLOCKED (default: 3)
+CB_THRESHOLD="${CB_THRESHOLD:-6}"           # consecutive failures before BLOCKED (default: 6)
 # Effective CB threshold: doubled when consensus mode active (AC2 auto-double)
 if [[ "${VERIFY_CONSENSUS:-0}" = "1" ]]; then
   EFFECTIVE_CB_THRESHOLD=$(( CB_THRESHOLD * 2 ))
@@ -192,7 +192,7 @@ launch_worker_codex() {
 
   log "  Launching Worker codex via trigger script in pane $pane_id..."
   paste_to_pane "$pane_id" "bash $trigger_file"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
   log_debug "Worker codex trigger sent: $trigger_file"
   sleep 3  # brief wait for codex to start
   return 0
@@ -211,7 +211,7 @@ launch_worker_claude() {
 
   log "  Launching Worker claude in pane $pane_id..."
   paste_to_pane "$pane_id" "$worker_launch"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
 
   # Wait for claude TUI to be ready
   if ! wait_for_pane_ready "$pane_id" 30; then
@@ -223,7 +223,7 @@ launch_worker_claude() {
   sleep 3
   local worker_instruction="Read and execute the instructions in $prompt_file"
   paste_to_pane "$pane_id" "$worker_instruction"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
   log_debug "Worker instruction sent directly (${#worker_instruction} chars)"
 
   # 15-iteration submit loop — verify claude started working
@@ -244,7 +244,7 @@ launch_worker_claude() {
       sleep 0.2
       paste_to_pane "$pane_id" "$worker_instruction"
       sleep 0.15
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       sleep 1
     fi
     tmux send-keys -t "$pane_id" C-m 2>/dev/null
@@ -259,15 +259,15 @@ launch_worker_claude() {
     log_debug "[GOV] iter=$iter worker_instruction_failed=true attempts=15 action=restart_claude"
     tmux send-keys -t "$pane_id" C-c 2>/dev/null
     sleep 0.5
-    tmux send-keys -t "$pane_id" "/exit" Enter 2>/dev/null
+    tmux send-keys -t "$pane_id" "/exit" C-m 2>/dev/null
     sleep 2
     wait_for_pane_ready "$pane_id" 10 2>/dev/null || true
     paste_to_pane "$pane_id" "$worker_launch"
-    tmux send-keys -t "$pane_id" Enter
+    tmux send-keys -t "$pane_id" C-m
     if wait_for_pane_ready "$pane_id" 30; then
       sleep 3
       paste_to_pane "$pane_id" "$worker_instruction"
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       log "  Worker restarted and instruction re-sent"
       log_debug "[FLOW] iter=$iter worker_restart_recovery=success"
     else
@@ -290,7 +290,7 @@ launch_verifier_codex() {
 
   log "  Launching Verifier codex in pane $pane_id..."
   paste_to_pane "$pane_id" "$verifier_launch"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
   sleep 3
   return 0
 }
@@ -306,7 +306,7 @@ launch_verifier_claude() {
 
   log "  Launching Verifier claude in pane $pane_id..."
   paste_to_pane "$pane_id" "$verifier_launch"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
 
   if ! wait_for_pane_ready "$pane_id" 30; then
     log_error "Verifier failed to start"
@@ -316,7 +316,7 @@ launch_verifier_claude() {
   sleep 3
   local verifier_instruction="Read and execute the instructions in $prompt_file"
   paste_to_pane "$pane_id" "$verifier_instruction"
-  tmux send-keys -t "$pane_id" Enter
+  tmux send-keys -t "$pane_id" C-m
   log_debug "Verifier instruction sent directly"
 
   # Submit loop — verify verifier started working
@@ -334,7 +334,7 @@ launch_verifier_claude() {
       tmux send-keys -t "$pane_id" C-u 2>/dev/null
       sleep 0.1
       paste_to_pane "$pane_id" "$verifier_instruction"
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
     fi
     tmux send-keys -t "$pane_id" C-m 2>/dev/null
     sleep 0.3
@@ -663,13 +663,13 @@ safe_send_keys() {
   # Auto-approve permission prompts ("Do you want to create/overwrite X?")
   if echo "$initial_capture" | grep -q "Do you want to" 2>/dev/null; then
     log_debug " Permission prompt detected, auto-approving"
-    tmux send-keys -t "$pane_id" Enter
+    tmux send-keys -t "$pane_id" C-m
     sleep 0.3
   fi
   # Auto-dismiss codex update prompt (select Skip)
   if echo "$initial_capture" | grep -qi "new version\|update.*codex\|codex.*update" 2>/dev/null; then
     log_debug " Codex update prompt detected, selecting Skip"
-    tmux send-keys -t "$pane_id" "2" Enter
+    tmux send-keys -t "$pane_id" "2" C-m
     sleep 0.2
   fi
   # Send text via buffer paste (reliable for long strings)
@@ -761,9 +761,9 @@ wait_for_pane_ready() {
     # Auto-dismiss trust prompt (tmux pattern: paneHasTrustPrompt)
     if echo "$captured" | grep -q "Do you trust" 2>/dev/null; then
       log "  Trust prompt detected, auto-dismissing..."
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       sleep 0.12
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       sleep 2
       continue
     fi
@@ -771,7 +771,7 @@ wait_for_pane_ready() {
     # Auto-approve permission prompts ("Do you want to create/overwrite X?")
     if echo "$captured" | grep -q "Do you want to" 2>/dev/null; then
       log "  Permission prompt detected, auto-approving..."
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       sleep 0.5
       continue
     fi
@@ -779,7 +779,7 @@ wait_for_pane_ready() {
     # Auto-dismiss codex update prompt (select Skip = option 2)
     if echo "$captured" | grep -qi "new version\|update.*codex\|codex.*update" 2>/dev/null; then
       log "  Codex update prompt detected, selecting Skip..."
-      tmux send-keys -t "$pane_id" "2" Enter
+      tmux send-keys -t "$pane_id" "2" C-m
       sleep 0.5
       continue
     fi
@@ -917,7 +917,7 @@ restart_worker() {
 
   # Kill existing claude, wait for shell prompt
   tmux send-keys -t "$pane_id" C-c 2>/dev/null
-  tmux send-keys -t "$pane_id" "/exit" Enter 2>/dev/null
+  tmux send-keys -t "$pane_id" "/exit" C-m 2>/dev/null
   sleep 2
 
   # Re-launch worker (tmux interactive pattern)
@@ -1205,11 +1205,11 @@ cleanup() {
   log_debug "cleanup: WORKER_PANE=${WORKER_PANE:-unset} VERIFIER_PANE=${VERIFIER_PANE:-unset}"
   if [[ -n "${WORKER_PANE:-}" ]]; then
     tmux send-keys -t "$WORKER_PANE" C-c 2>/dev/null
-    tmux send-keys -t "$WORKER_PANE" "/exit" Enter 2>/dev/null
+    tmux send-keys -t "$WORKER_PANE" "/exit" C-m 2>/dev/null
   fi
   if [[ -n "${VERIFIER_PANE:-}" ]]; then
     tmux send-keys -t "$VERIFIER_PANE" C-c 2>/dev/null
-    tmux send-keys -t "$VERIFIER_PANE" "/exit" Enter 2>/dev/null
+    tmux send-keys -t "$VERIFIER_PANE" "/exit" C-m 2>/dev/null
   fi
   sleep 2
   # Kill panes on completion
@@ -1410,7 +1410,7 @@ poll_for_signal() {
         log "  A5: Rate-limited pane shows 'queued messages' — restarting $role pane"
         log_debug "[GOV] iter=$ITERATION phase=rate_limit_pane_restart role=$role reason=queued_messages"
         tmux send-keys -t "$pane_id" C-c 2>/dev/null; sleep 0.5
-        tmux send-keys -t "$pane_id" "/exit" Enter 2>/dev/null; sleep 2
+        tmux send-keys -t "$pane_id" "/exit" C-m 2>/dev/null; sleep 2
         wait_for_pane_ready "$pane_id" 10 2>/dev/null || true
       fi
       sleep "$_API_RETRY_INTERVAL_S"
@@ -1487,7 +1487,7 @@ poll_for_signal() {
     if echo "$poll_capture" | grep -q "Do you want to" 2>/dev/null; then
       log "  Permission prompt detected during poll, auto-approving..."
       log_debug "[FLOW] iter=$ITERATION permission_prompt_auto_approved=true"
-      tmux send-keys -t "$pane_id" Enter
+      tmux send-keys -t "$pane_id" C-m
       sleep 0.5
     fi
 
@@ -1529,12 +1529,12 @@ run_single_verifier() {
     log_debug "[GOV] iter=$iter pane_dead=true pane_id=$VERIFIER_PANE cmd=$verifier_cmd action=reset_shell"
     tmux send-keys -t "$VERIFIER_PANE" C-c C-u 2>/dev/null
     sleep 0.2
-    tmux send-keys -t "$VERIFIER_PANE" "clear" Enter 2>/dev/null
+    tmux send-keys -t "$VERIFIER_PANE" "clear" C-m 2>/dev/null
     sleep 0.3
   elif [[ "$verifier_cmd" == "node" || "$verifier_cmd" == "claude" || "$verifier_cmd" == "codex" ]]; then
     tmux send-keys -t "$VERIFIER_PANE" C-c 2>/dev/null
     sleep 0.5
-    tmux send-keys -t "$VERIFIER_PANE" "/exit" Enter 2>/dev/null
+    tmux send-keys -t "$VERIFIER_PANE" "/exit" C-m 2>/dev/null
     sleep 2
   fi
   # Always ensure clean shell state before launching new verifier
@@ -1628,7 +1628,7 @@ run_sequential_final_verify() {
     verifier_cmd=$(tmux display-message -p -t "$VERIFIER_PANE" '#{pane_current_command}' 2>/dev/null)
     if [[ "$verifier_cmd" == "node" || "$verifier_cmd" == "claude" || "$verifier_cmd" == "codex" ]]; then
       tmux send-keys -t "$VERIFIER_PANE" C-c 2>/dev/null; sleep 0.5
-      tmux send-keys -t "$VERIFIER_PANE" "/exit" Enter 2>/dev/null; sleep 2
+      tmux send-keys -t "$VERIFIER_PANE" "/exit" C-m 2>/dev/null; sleep 2
     fi
     wait_for_pane_ready "$VERIFIER_PANE" 10 2>/dev/null || true
 
@@ -1887,12 +1887,10 @@ main() {
   trap cleanup EXIT INT TERM
   mkdir -p "$LOGS_DIR" "$RUNTIME_DIR" 2>/dev/null
 
-  # --- Analytics directory: create only when --debug or --with-self-verification ---
-  if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
-    mkdir -p "$ANALYTICS_DIR" 2>/dev/null
-  fi
+  # --- Analytics directory: always create (campaign.jsonl + metadata.json are always-on) ---
+  mkdir -p "$ANALYTICS_DIR" 2>/dev/null
 
-  # --- debug.log versioning (in analytics dir) ---
+  # --- debug.log versioning (in analytics dir, --debug only) ---
   if (( DEBUG )) && [[ -f "$DEBUG_LOG" ]]; then
     local dbg_n=1
     while [[ -f "${DEBUG_LOG%.log}-v${dbg_n}.log" ]]; do
@@ -1901,33 +1899,30 @@ main() {
     mv "$DEBUG_LOG" "${DEBUG_LOG%.log}-v${dbg_n}.log"
   fi
 
-  # --- campaign.jsonl versioning (in analytics dir, after mkdir) ---
-  if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
-    if [[ -f "$CAMPAIGN_JSONL" ]]; then
-      local cj_n=1
-      while [[ -f "${CAMPAIGN_JSONL%.jsonl}-v${cj_n}.jsonl" ]]; do
-        (( cj_n++ ))
-      done
-      mv "$CAMPAIGN_JSONL" "${CAMPAIGN_JSONL%.jsonl}-v${cj_n}.jsonl"
-    fi
+  # --- campaign.jsonl versioning (always-on) ---
+  if [[ -f "$CAMPAIGN_JSONL" ]]; then
+    local cj_n=1
+    while [[ -f "${CAMPAIGN_JSONL%.jsonl}-v${cj_n}.jsonl" ]]; do
+      (( cj_n++ ))
+    done
+    mv "$CAMPAIGN_JSONL" "${CAMPAIGN_JSONL%.jsonl}-v${cj_n}.jsonl"
   fi
 
-  # --- metadata.json: write at campaign start ---
-  if (( DEBUG )) || (( WITH_SELF_VERIFICATION )); then
-    jq -n \
-      --arg slug "$SLUG" \
-      --arg project_root "$ROOT" \
-      --arg campaign_status "running" \
-      --arg start_time "$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
-      --arg end_time "" \
-      --arg worker_model "$WORKER_MODEL" \
-      --arg verifier_model "$VERIFIER_MODEL" \
-      --argjson debug "$DEBUG" \
-      --argjson with_sv "$WITH_SELF_VERIFICATION" \
-      --argjson consensus "$VERIFY_CONSENSUS" \
-      '{slug: $slug, project_root: $project_root, campaign_status: $campaign_status, start_time: $start_time, end_time: $end_time, worker_model: $worker_model, verifier_model: $verifier_model, debug: $debug, with_self_verification: $with_sv, consensus: $consensus}' \
-      > "$METADATA_FILE"
-  fi
+  # --- metadata.json: always write at campaign start (cross-project identification) ---
+  jq -n \
+    --arg slug "$SLUG" \
+    --arg project_root "$ROOT" \
+    --arg project_name "$(basename "$ROOT")" \
+    --arg campaign_status "running" \
+    --arg start_time "$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
+    --arg end_time "" \
+    --arg worker_model "$WORKER_MODEL" \
+    --arg verifier_model "$VERIFIER_MODEL" \
+    --argjson debug "$DEBUG" \
+    --argjson with_sv "$WITH_SELF_VERIFICATION" \
+    --argjson consensus "$VERIFY_CONSENSUS" \
+    '{slug: $slug, project_root: $project_root, project_name: $project_name, campaign_status: $campaign_status, start_time: $start_time, end_time: $end_time, worker_model: $worker_model, verifier_model: $verifier_model, debug: $debug, with_self_verification: $with_sv, consensus: $consensus}' \
+    > "$METADATA_FILE"
 
   # --- Startup ---
   log "Ralph Desk Tmux Runner starting..."
@@ -2065,7 +2060,7 @@ main() {
       # Send C-c first (in case claude is mid-task), then /exit
       tmux send-keys -t "$WORKER_PANE" C-c 2>/dev/null
       sleep 1
-      tmux send-keys -t "$WORKER_PANE" "/exit" Enter 2>/dev/null
+      tmux send-keys -t "$WORKER_PANE" "/exit" C-m 2>/dev/null
       sleep 2
       # Wait for shell prompt before proceeding
       wait_for_pane_ready "$WORKER_PANE" 10 2>/dev/null || true
@@ -2261,12 +2256,12 @@ main() {
             log_debug "[GOV] iter=$ITERATION pane_dead=true pane_id=$VERIFIER_PANE cmd=$verifier_cmd action=reset_shell"
             tmux send-keys -t "$VERIFIER_PANE" C-c C-u 2>/dev/null
             sleep 0.2
-            tmux send-keys -t "$VERIFIER_PANE" "clear" Enter 2>/dev/null
+            tmux send-keys -t "$VERIFIER_PANE" "clear" C-m 2>/dev/null
             sleep 0.3
           elif [[ "$verifier_cmd" == "node" || "$verifier_cmd" == "claude" || "$verifier_cmd" == "codex" ]]; then
             tmux send-keys -t "$VERIFIER_PANE" C-c 2>/dev/null
             sleep 0.5
-            tmux send-keys -t "$VERIFIER_PANE" "/exit" Enter 2>/dev/null
+            tmux send-keys -t "$VERIFIER_PANE" "/exit" C-m 2>/dev/null
             sleep 2
           fi
           wait_for_pane_ready "$VERIFIER_PANE" 10 2>/dev/null || true