@the-bearded-bear/claude-craft 9.2.1 → 9.3.0-next.a0b7093

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (192) hide show
  1. package/Dev/i18n/base/Common/hooks/scripts/block-dangerous-commands.sh +1 -1
  2. package/Dev/i18n/base/Common/hooks/scripts/guard-network.sh +1 -1
  3. package/Dev/i18n/base/Common/hooks/scripts/notify-slack.sh +1 -1
  4. package/Dev/i18n/base/Common/hooks/scripts/output-filter.sh +2 -2
  5. package/Dev/i18n/base/Common/hooks/scripts/output-size-hint.sh +3 -3
  6. package/Dev/i18n/base/Common/hooks/scripts/post-edit-lint.sh +3 -3
  7. package/Dev/i18n/base/Common/hooks/scripts/post-tool-failure.sh +2 -2
  8. package/Dev/i18n/base/Common/hooks/scripts/pre-commit-check.sh +7 -7
  9. package/Dev/i18n/base/Common/hooks/scripts/pre-compact.sh +8 -8
  10. package/Dev/i18n/base/Common/hooks/scripts/protect-sensitive-files.sh +2 -2
  11. package/Dev/i18n/base/Common/hooks/scripts/quality-gate.sh +10 -10
  12. package/Dev/i18n/base/Common/hooks/scripts/session-end.sh +3 -3
  13. package/Dev/i18n/base/Common/hooks/scripts/session-init.sh +8 -8
  14. package/Dev/i18n/base/Common/hooks/templates/examples/block-dangerous-commands.json +2 -1
  15. package/Dev/i18n/base/Common/hooks/templates/examples/protect-files.json +2 -1
  16. package/Dev/i18n/base/Common/hooks/templates/examples/security-block.json +2 -1
  17. package/Dev/i18n/base/Common/hooks/templates/hooks.default.json +6 -3
  18. package/Dev/i18n/base/Common/skills/ecosystem-tools/SKILL.md +1 -1
  19. package/Dev/i18n/de/Common/hooks/HOOKS-QUICKSTART.md +1 -0
  20. package/Dev/i18n/de/Common/rules/12-context-management.md +6 -4
  21. package/Dev/i18n/de/Workflow/commands/start.md +1 -1
  22. package/Dev/i18n/en/Common/hooks/HOOKS-QUICKSTART.md +1 -0
  23. package/Dev/i18n/en/Common/rules/12-context-management.md +6 -4
  24. package/Dev/i18n/en/Workflow/commands/start.md +1 -1
  25. package/Dev/i18n/es/Common/hooks/HOOKS-QUICKSTART.md +1 -0
  26. package/Dev/i18n/es/Common/rules/12-context-management.md +6 -4
  27. package/Dev/i18n/es/Workflow/commands/start.md +1 -1
  28. package/Dev/i18n/fr/Common/hooks/HOOKS-QUICKSTART.md +1 -0
  29. package/Dev/i18n/fr/Common/rules/12-context-management.md +6 -4
  30. package/Dev/i18n/fr/Workflow/commands/start.md +1 -1
  31. package/Dev/i18n/pt/Common/hooks/HOOKS-QUICKSTART.md +1 -0
  32. package/Dev/i18n/pt/Common/rules/12-context-management.md +6 -4
  33. package/Dev/i18n/pt/Workflow/commands/start.md +1 -1
  34. package/Dev/scripts/lib/install-tech-common.sh +39 -39
  35. package/Dev/scripts/lib/merge-hooks.mjs +20 -9
  36. package/Dev/scripts/tcl-common.sh +23 -23
  37. package/Infra/install-ansible-rules.sh +26 -26
  38. package/Infra/install-coolify-rules.sh +26 -26
  39. package/Infra/install-docker-rules.sh +26 -26
  40. package/Infra/install-frankenphp-rules.sh +26 -26
  41. package/Infra/install-hcloud-rules.sh +26 -26
  42. package/Infra/install-kubernetes-rules.sh +26 -26
  43. package/Infra/install-opentofu-rules.sh +26 -26
  44. package/Infra/install-pgbouncer-rules.sh +26 -26
  45. package/Project/i18n/de/templates/sprint-status.md +4 -0
  46. package/Project/i18n/en/templates/sprint-status.md +4 -0
  47. package/Project/i18n/es/templates/sprint-status.md +4 -0
  48. package/Project/i18n/fr/templates/sprint-status.md +4 -0
  49. package/Project/i18n/pt/templates/sprint-status.md +4 -0
  50. package/Project/install-project-commands.sh +9 -9
  51. package/Project/templates/sprint-status.yaml.template +4 -0
  52. package/README.md +15 -3
  53. package/Tools/ProjectConfig/claude-projects.sh +2 -2
  54. package/Tools/RTK/install-rtk.sh +1 -1
  55. package/Tools/Ralph/hooks/lib/ralph-common.sh +3 -3
  56. package/Tools/Ralph/hooks/pre-tool-context.sh +2 -2
  57. package/Tools/Ralph/hooks/session-restore.sh +8 -8
  58. package/Tools/Ralph/hooks/stop-dod-gate.sh +6 -6
  59. package/Tools/Ralph/lib/metrics-exporter.sh +3 -3
  60. package/Tools/Ralph/lib/parallel-manager.sh +1 -1
  61. package/Tools/Ralph/lib/recovery-engine.sh +2 -1
  62. package/Tools/Ralph/lib/sprint-conductor.sh +20 -7
  63. package/bundles/cursor/.cursorrules +2 -2
  64. package/bundles/windsurf/.windsurfrules +2 -2
  65. package/cli/kanban/client/dist/assets/{BurndownView-B6D7QRcK.css → BurndownView-CnsZ1jie.css} +1 -1
  66. package/cli/kanban/client/dist/assets/BurndownView-s2wpjiqE.js +3 -0
  67. package/cli/kanban/client/dist/assets/DepsView-Bo8sRS7i.js +1 -0
  68. package/cli/kanban/client/dist/assets/DocsView-QjjjfhpV.js +1 -0
  69. package/cli/kanban/client/dist/assets/SprintsView-B4toxGc4.css +1 -0
  70. package/cli/kanban/client/dist/assets/SprintsView-BSh9XshU.js +1 -0
  71. package/cli/kanban/client/dist/assets/actions-D0g4C88h.js +2 -0
  72. package/cli/kanban/client/dist/assets/de-CpCArOVg.js +1 -0
  73. package/cli/kanban/client/dist/assets/es-CnARpMvm.js +1 -0
  74. package/cli/kanban/client/dist/assets/fr-CIeAvktE.js +1 -0
  75. package/cli/kanban/client/dist/assets/index-BimcLpOz.css +1 -0
  76. package/cli/kanban/client/dist/assets/index-CdlYlxD5.js +2 -0
  77. package/cli/kanban/client/dist/assets/palette.svelte-BymgBiMI.js +1 -0
  78. package/cli/kanban/client/dist/assets/pt-Cs89lHuV.js +1 -0
  79. package/cli/kanban/client/dist/assets/router.svelte-CKx0fx-k.js +1 -0
  80. package/cli/kanban/client/dist/index.html +4 -2
  81. package/cli/kanban/client/src/App.svelte +212 -105
  82. package/cli/kanban/client/src/app.css +234 -62
  83. package/cli/kanban/client/src/components/Avatar.svelte +6 -4
  84. package/cli/kanban/client/src/components/CapacityBar.svelte +57 -0
  85. package/cli/kanban/client/src/components/CloseSprintDialog.svelte +114 -0
  86. package/cli/kanban/client/src/components/ConfirmDialog.svelte +125 -0
  87. package/cli/kanban/client/src/components/FilterBar.svelte +179 -0
  88. package/cli/kanban/client/src/components/KeyboardShortcuts.svelte +16 -0
  89. package/cli/kanban/client/src/components/MoveMenu.svelte +42 -0
  90. package/cli/kanban/client/src/components/NotFound.svelte +10 -0
  91. package/cli/kanban/client/src/components/Points.svelte +8 -2
  92. package/cli/kanban/client/src/components/PriorityChip.svelte +4 -1
  93. package/cli/kanban/client/src/components/ProgressRing.svelte +7 -4
  94. package/cli/kanban/client/src/components/PromptDialog.svelte +95 -54
  95. package/cli/kanban/client/src/components/SprintForm.svelte +189 -0
  96. package/cli/kanban/client/src/components/StoryCreateDialog.svelte +133 -0
  97. package/cli/kanban/client/src/components/StoryDetail.svelte +744 -76
  98. package/cli/kanban/client/src/components/StoryForm.svelte +340 -0
  99. package/cli/kanban/client/src/components/StoryRow.svelte +45 -0
  100. package/cli/kanban/client/src/components/TaskProgress.svelte +4 -2
  101. package/cli/kanban/client/src/components/TddPip.svelte +10 -8
  102. package/cli/kanban/client/src/components/Toasts.svelte +52 -0
  103. package/cli/kanban/client/src/components/TweaksPanel.svelte +86 -32
  104. package/cli/kanban/client/src/components/WipSettings.svelte +130 -0
  105. package/cli/kanban/client/src/lib/actions.js +49 -0
  106. package/cli/kanban/client/src/lib/announce.js +28 -0
  107. package/cli/kanban/client/src/lib/api.js +77 -0
  108. package/cli/kanban/client/src/lib/backlog.js +38 -4
  109. package/cli/kanban/client/src/lib/board-card.js +50 -0
  110. package/cli/kanban/client/src/lib/board-nav.js +111 -0
  111. package/cli/kanban/client/src/lib/burndown.js +15 -0
  112. package/cli/kanban/client/src/lib/config.js +8 -0
  113. package/cli/kanban/client/src/lib/deps-style.js +5 -2
  114. package/cli/kanban/client/src/lib/deps-table.js +51 -0
  115. package/cli/kanban/client/src/lib/detail.svelte.js +31 -23
  116. package/cli/kanban/client/src/lib/docs-tree.js +30 -0
  117. package/cli/kanban/client/src/lib/events.js +73 -8
  118. package/cli/kanban/client/src/lib/filter.js +68 -0
  119. package/cli/kanban/client/src/lib/health.svelte.js +19 -0
  120. package/cli/kanban/client/src/lib/i18n/index.svelte.js +73 -0
  121. package/cli/kanban/client/src/lib/i18n/locales/de.js +587 -0
  122. package/cli/kanban/client/src/lib/i18n/locales/en.js +588 -0
  123. package/cli/kanban/client/src/lib/i18n/locales/es.js +584 -0
  124. package/cli/kanban/client/src/lib/i18n/locales/fr.js +584 -0
  125. package/cli/kanban/client/src/lib/i18n/locales/pt.js +583 -0
  126. package/cli/kanban/client/src/lib/i18n/translate.js +86 -0
  127. package/cli/kanban/client/src/lib/icons.js +0 -5
  128. package/cli/kanban/client/src/lib/leave-guard.js +29 -0
  129. package/cli/kanban/client/src/lib/markdown.js +99 -0
  130. package/cli/kanban/client/src/lib/opener.js +12 -0
  131. package/cli/kanban/client/src/lib/optimistic.js +44 -0
  132. package/cli/kanban/client/src/lib/palette.svelte.js +10 -0
  133. package/cli/kanban/client/src/lib/points.js +6 -0
  134. package/cli/kanban/client/src/lib/query.js +92 -0
  135. package/cli/kanban/client/src/lib/router.svelte.js +70 -11
  136. package/cli/kanban/client/src/lib/shell.js +56 -0
  137. package/cli/kanban/client/src/lib/sprint-actions.svelte.js +151 -0
  138. package/cli/kanban/client/src/lib/sprint-plan.js +141 -0
  139. package/cli/kanban/client/src/lib/sprints.js +38 -1
  140. package/cli/kanban/client/src/lib/store.svelte.js +195 -78
  141. package/cli/kanban/client/src/lib/story-actions.svelte.js +236 -0
  142. package/cli/kanban/client/src/lib/story-form.js +205 -0
  143. package/cli/kanban/client/src/lib/toasts.js +84 -0
  144. package/cli/kanban/client/src/lib/tweaks.svelte.js +6 -9
  145. package/cli/kanban/client/src/lib/velocity.js +35 -0
  146. package/cli/kanban/client/src/lib/wip.js +17 -0
  147. package/cli/kanban/client/src/views/BacklogView.svelte +206 -91
  148. package/cli/kanban/client/src/views/BurndownView.svelte +203 -104
  149. package/cli/kanban/client/src/views/DepsView.svelte +176 -100
  150. package/cli/kanban/client/src/views/DocsView.svelte +170 -122
  151. package/cli/kanban/client/src/views/KanbanView.svelte +377 -356
  152. package/cli/kanban/client/src/views/SprintsView.svelte +313 -179
  153. package/cli/kanban/client/src/views.css +1518 -305
  154. package/cli/kanban/server/app.js +38 -241
  155. package/cli/kanban/server/cli-options.js +58 -0
  156. package/cli/kanban/server/middleware/security.js +28 -2
  157. package/cli/kanban/server/routes/http-errors.js +55 -0
  158. package/cli/kanban/server/routes/settings.js +24 -0
  159. package/cli/kanban/server/routes/sprints.js +364 -0
  160. package/cli/kanban/server/routes/stories.js +565 -0
  161. package/cli/kanban/server/services/event-bus.js +3 -2
  162. package/cli/kanban/server/services/file-scanner.js +2 -1
  163. package/cli/kanban/server/services/file-watcher.js +27 -10
  164. package/cli/kanban/server/services/file-writer.js +12 -58
  165. package/cli/kanban/server/services/frontmatter.js +44 -6
  166. package/cli/kanban/server/services/idempotency.js +77 -0
  167. package/cli/kanban/server/services/kanban-config.js +122 -0
  168. package/cli/kanban/server/services/lock.js +138 -0
  169. package/cli/kanban/server/services/markdown-metadata.js +27 -13
  170. package/cli/kanban/server/services/repository.js +268 -165
  171. package/cli/kanban/server/services/sprint-cache.js +24 -136
  172. package/cli/kanban/server/services/sprint-ops.js +223 -0
  173. package/cli/kanban/server/services/state-machine.js +8 -11
  174. package/cli/kanban/server/services/status-store.js +385 -0
  175. package/cli/kanban/server/services/story-files.js +282 -0
  176. package/cli/kanban/shared/ids.js +42 -0
  177. package/cli/kanban/shared/schemas.js +106 -12
  178. package/cli/kanban/shared/transitions.js +25 -0
  179. package/cli/lib/doctor.js +8 -7
  180. package/cli/lib/installer.js +2 -2
  181. package/cli/lib/kanban.js +11 -7
  182. package/package.json +2 -1
  183. package/cli/kanban/client/dist/assets/BurndownView-BBl1LOfu.js +0 -3
  184. package/cli/kanban/client/dist/assets/DepsView-D0N9Tg4d.css +0 -1
  185. package/cli/kanban/client/dist/assets/DepsView-DXREli3v.js +0 -1
  186. package/cli/kanban/client/dist/assets/DocsView--bhcgCS3.js +0 -1
  187. package/cli/kanban/client/dist/assets/DocsView-CBBzt7Ru.css +0 -1
  188. package/cli/kanban/client/dist/assets/SprintsView-8AMci9M2.css +0 -1
  189. package/cli/kanban/client/dist/assets/SprintsView-CLKf2BTm.js +0 -1
  190. package/cli/kanban/client/dist/assets/index-DI89vrQ2.js +0 -4
  191. package/cli/kanban/client/dist/assets/index-DRj2q7mf.css +0 -1
  192. package/cli/kanban/client/src/lib/kanban-nav.js +0 -119
@@ -16,7 +16,7 @@ set -u
16
16
  hook_require_parser
17
17
  hook_input
18
18
  CMD=$(hook_get tool_input.command)
19
- [ -n "$CMD" ] || exit 0
19
+ [[ -n "$CMD" ]] || exit 0
20
20
 
21
21
  MODE=${BLOCK_RM_RF:-dangerous}
22
22
  TEXT=$(printf '%s\n' "$CMD" | hook_strip_heredocs | hook_shell_segments)
@@ -13,7 +13,7 @@ set -u
13
13
  hook_require_parser
14
14
  hook_input
15
15
  CMD=$(hook_get tool_input.command)
16
- [ -n "$CMD" ] || exit 0
16
+ [[ -n "$CMD" ]] || exit 0
17
17
 
18
18
  # Corps de heredoc retirés, puis un segment par ligne (hook_shell_segments) : le texte
19
19
  # cité est une donnée, sauf après sh -c / eval et dans $(…) / `…`. Un segment "|"
@@ -7,7 +7,7 @@ set -u
7
7
  # shellcheck source=lib/hook-io.sh
8
8
  . "$(dirname "${BASH_SOURCE[0]}")/lib/hook-io.sh"
9
9
 
10
- [ -n "${SLACK_WEBHOOK_URL:-}" ] || exit 0
10
+ [[ -n "${SLACK_WEBHOOK_URL:-}" ]] || exit 0
11
11
  hook_has_parser || exit 0
12
12
  hook_input
13
13
  MESSAGE=$(hook_get message)
@@ -11,10 +11,10 @@ hook_has_parser || exit 0
11
11
  hook_input
12
12
  OUT=$(hook_get tool_response.stdout)
13
13
  LEN=${#OUT}
14
- [ "$LEN" -gt 50000 ] || exit 0
14
+ [[ "$LEN" -gt 50000 ]] || exit 0
15
15
 
16
16
  INTERRUPTED=false
17
- [ "$(hook_get tool_response.interrupted)" != true ] || INTERRUPTED=true
17
+ [[ "$(hook_get tool_response.interrupted)" != true ]] || INTERRUPTED=true
18
18
  HEAD=$(printf '%s' "$OUT" | head -c 8000)
19
19
  hook_replace_bash_output \
20
20
  "[output truncated by the output-filter hook: ${LEN} chars total, first 8000 shown. Re-run with a narrower command (grep/head/tail) to see more.]
@@ -8,12 +8,12 @@ set -u
8
8
  hook_has_parser || exit 0
9
9
  hook_input
10
10
  OUT=$(hook_get tool_response.stdout)
11
- [ -n "$OUT" ] || OUT=$(hook_get tool_response)
11
+ [[ -n "$OUT" ]] || OUT=$(hook_get tool_response)
12
12
  LEN=${#OUT}
13
13
 
14
- if [ "$LEN" -gt 50000 ]; then
14
+ if [[ "$LEN" -gt 50000 ]]; then
15
15
  hook_context PostToolUse "Output very large (${LEN} chars). Summarize key findings only; do not repeat output verbatim."
16
- elif [ "$(hook_get tool_name)" = "Bash" ] && [ "$LEN" -gt 10000 ]; then
16
+ elif [[ "$(hook_get tool_name)" = "Bash" ]] && [[ "$LEN" -gt 10000 ]]; then
17
17
  hook_context PostToolUse "Output: $(printf '%s\n' "$OUT" | wc -l | tr -d ' ') lines. Extract only relevant information."
18
18
  fi
19
19
  exit 0
@@ -8,18 +8,18 @@ set -u
8
8
  hook_has_parser || exit 0
9
9
  hook_input
10
10
  FILE_PATH=$(hook_get tool_input.file_path)
11
- [ -n "$FILE_PATH" ] || exit 0
11
+ [[ -n "$FILE_PATH" ]] || exit 0
12
12
 
13
13
  have() { command -v "$1" >/dev/null 2>&1; }
14
14
 
15
15
  case "${FILE_PATH##*.}" in
16
16
  js | jsx | ts | tsx)
17
- if have npx && [ -f package.json ]; then npx --no-install eslint --fix "$FILE_PATH" >/dev/null 2>&1 || true; fi
17
+ if have npx && [[ -f package.json ]]; then npx --no-install eslint --fix "$FILE_PATH" >/dev/null 2>&1 || true; fi
18
18
  ;;
19
19
  php)
20
20
  if have php-cs-fixer; then
21
21
  php-cs-fixer fix "$FILE_PATH" --quiet >/dev/null 2>&1 || true
22
- elif [ -x vendor/bin/php-cs-fixer ]; then
22
+ elif [[ -x vendor/bin/php-cs-fixer ]]; then
23
23
  vendor/bin/php-cs-fixer fix "$FILE_PATH" --quiet >/dev/null 2>&1 || true
24
24
  fi
25
25
  ;;
@@ -12,7 +12,7 @@ hook_input
12
12
  TOOL=$(hook_get tool_name)
13
13
  ERROR=$(hook_get error)
14
14
  CWD=$(hook_get cwd)
15
- [ -z "$CWD" ] || cd "$CWD" 2>/dev/null || true
15
+ [[ -z "$CWD" ]] || cd "$CWD" 2>/dev/null || true
16
16
 
17
17
  has() { printf '%s' "$ERROR" | grep -qiE -e "$1"; }
18
18
 
@@ -46,5 +46,5 @@ printf '[%s] [%s] [%s] %s\n' "$(date -u +%Y-%m-%dT%H:%M:%SZ)" "$severity" "$TOOL
46
46
  "$(printf '%s' "$ERROR" | tr '\n' ' ' | cut -c1-300)" >>.claude/logs/tool-failures.log 2>/dev/null
47
47
 
48
48
  HINT=$(recovery_hint)
49
- [ -z "$HINT" ] || hook_context PostToolUseFailure "$HINT"
49
+ [[ -z "$HINT" ]] || hook_context PostToolUseFailure "$HINT"
50
50
  exit 0
@@ -12,9 +12,9 @@ set -u
12
12
  hook_require_parser
13
13
  hook_input
14
14
  CMD=$(hook_get tool_input.command)
15
- [ -n "$CMD" ] || exit 0
15
+ [[ -n "$CMD" ]] || exit 0
16
16
  cwd=$(hook_get cwd)
17
- [ -z "$cwd" ] || cd "$cwd" 2>/dev/null || exit 0
17
+ [[ -z "$cwd" ]] || cd "$cwd" 2>/dev/null || exit 0
18
18
 
19
19
  # Lignes émises : `commit`, `all` (commit -a), `add` (git add), `dir <chemin>` (git -C).
20
20
  facts=$(printf '%s\n' "$CMD" | hook_strip_heredocs | tr "'\"" ' ' | awk '
@@ -41,7 +41,7 @@ facts=$(printf '%s\n' "$CMD" | hook_strip_heredocs | tr "'\"" ' ' | awk '
41
41
  case "$facts" in *commit*) ;; *) exit 0 ;; esac
42
42
 
43
43
  dir=$(printf '%s\n' "$facts" | sed -n 's/^dir //p' | head -n 1)
44
- [ -z "$dir" ] || cd "$dir" 2>/dev/null || exit 0
44
+ [[ -z "$dir" ]] || cd "$dir" 2>/dev/null || exit 0
45
45
  top=$(git rev-parse --show-toplevel 2>/dev/null) || exit 0
46
46
  cd "$top" || exit 0
47
47
 
@@ -50,19 +50,19 @@ if printf '%s\n' "$facts" | grep -qxE 'add|all'; then
50
50
  files=$(printf '%s\n%s\n%s\n' "$files" "$(git diff --name-only HEAD 2>/dev/null)" "$(git ls-files -o --exclude-standard 2>/dev/null)")
51
51
  fi
52
52
  files=$(printf '%s\n' "$files" | sed '/^$/d' | sort -u)
53
- [ -n "$files" ] || exit 0
53
+ [[ -n "$files" ]] || exit 0
54
54
 
55
55
  existing=()
56
56
  while IFS= read -r f; do
57
57
  if is_sensitive_path "$f"; then hook_block "Attempting to commit a sensitive file: $f"; fi
58
- [ -f "$f" ] && existing+=("$f")
58
+ [[ -f "$f" ]] && existing+=("$f")
59
59
  done <<EOF
60
60
  $files
61
61
  EOF
62
- [ ${#existing[@]} -gt 0 ] || exit 0
62
+ [[ ${#existing[@]} -gt 0 ]] || exit 0
63
63
 
64
64
  for p in '-----BEGIN [A-Z ]*PRIVATE KEY-----' 'AKIA[0-9A-Z]{16}' 'ghp_[A-Za-z0-9]{36}' 'xox[baprs]-[A-Za-z0-9-]{10,}'; do
65
65
  hit=$(grep -lIE -e "$p" -- "${existing[@]}" 2>/dev/null | head -n 1)
66
- [ -z "$hit" ] || hook_block "Potential secret detected in $hit"
66
+ [[ -z "$hit" ]] || hook_block "Potential secret detected in $hit"
67
67
  done
68
68
  exit 0
@@ -11,7 +11,7 @@ set -u
11
11
  hook_has_parser || exit 0
12
12
  hook_input
13
13
  CWD=$(hook_get cwd)
14
- [ -z "$CWD" ] || cd "$CWD" 2>/dev/null || true
14
+ [[ -z "$CWD" ]] || cd "$CWD" 2>/dev/null || true
15
15
  SESSION_ID=$(hook_get session_id)
16
16
  NOW=$(date -u +%Y-%m-%dT%H:%M:%SZ)
17
17
  STAMP=$(date -u +%Y%m%d%H%M%S)
@@ -28,25 +28,25 @@ prune() {
28
28
  local f i=0
29
29
  while IFS= read -r f; do
30
30
  i=$((i + 1))
31
- [ "$i" -le 10 ] || rm -f "$f"
31
+ [[ "$i" -le 10 ]] || rm -f "$f"
32
32
  done < <(ls -t "$1"/*.yaml 2>/dev/null)
33
33
  }
34
34
 
35
35
  log "Context compaction triggered (session: ${SESSION_ID:-unknown})"
36
36
  SAVED=""
37
- if [ -f .bmad/sprint-status.yaml ]; then
37
+ if [[ -f .bmad/sprint-status.yaml ]]; then
38
38
  backup .bmad/sprint-status.yaml .bmad/backups sprint-status && SAVED="$SAVED .bmad/sprint-status.yaml"
39
39
  fi
40
- if [ -f .bmad/current-story.yaml ]; then
40
+ if [[ -f .bmad/current-story.yaml ]]; then
41
41
  backup .bmad/current-story.yaml .bmad/backups current-story && SAVED="$SAVED .bmad/current-story.yaml"
42
42
  fi
43
- if [ -d .recette/sessions ]; then
43
+ if [[ -d .recette/sessions ]]; then
44
44
  ACTIVE=$(find .recette/sessions -name '*.yaml' -mmin -60 2>/dev/null | head -n 1)
45
- [ -z "$ACTIVE" ] || backup "$ACTIVE" .recette/backups "$(basename "$ACTIVE" .yaml)"
45
+ [[ -z "$ACTIVE" ]] || backup "$ACTIVE" .recette/backups "$(basename "$ACTIVE" .yaml)"
46
46
  fi
47
47
  for d in .bmad/backups .recette/backups; do
48
- [ ! -d "$d" ] || prune "$d"
48
+ [[ ! -d "$d" ]] || prune "$d"
49
49
  done
50
50
 
51
- [ -z "$SAVED" ] || printf 'When compacting, keep the current BMAD sprint and story context (backed up:%s).\n' "$SAVED"
51
+ [[ -z "$SAVED" ]] || printf 'When compacting, keep the current BMAD sprint and story context (backed up:%s).\n' "$SAVED"
52
52
  exit 0
@@ -8,9 +8,9 @@ set -u
8
8
  hook_require_parser
9
9
  hook_input
10
10
  FILE=$(hook_get tool_input.file_path)
11
- [ -n "$FILE" ] || FILE=$(hook_get tool_input.notebook_path)
11
+ [[ -n "$FILE" ]] || FILE=$(hook_get tool_input.notebook_path)
12
12
 
13
- if [ -n "$FILE" ] && is_sensitive_path "$FILE"; then
13
+ if [[ -n "$FILE" ]] && is_sensitive_path "$FILE"; then
14
14
  hook_block "Cannot modify sensitive file: $FILE"
15
15
  fi
16
16
  exit 0
@@ -11,29 +11,29 @@ set -u
11
11
 
12
12
  hook_has_parser || exit 0
13
13
  hook_input
14
- [ "$(hook_get stop_hook_active)" = "true" ] && exit 0
14
+ [[ "$(hook_get stop_hook_active)" = "true" ]] && exit 0
15
15
  cwd=$(hook_get cwd)
16
- [ -z "$cwd" ] || cd "$cwd" 2>/dev/null || exit 0
16
+ [[ -z "$cwd" ]] || cd "$cwd" 2>/dev/null || exit 0
17
17
 
18
18
  # Commande de test du projet (vide si aucun framework détecté).
19
19
  detect_cmd() {
20
- if [ -n "${QUALITY_GATE_CMD:-}" ]; then printf '%s' "$QUALITY_GATE_CMD"; return; fi
21
- if [ -f package.json ]; then
20
+ if [[ -n "${QUALITY_GATE_CMD:-}" ]]; then printf '%s' "$QUALITY_GATE_CMD"; return; fi
21
+ if [[ -f package.json ]]; then
22
22
  local saved=$HOOK_INPUT script
23
23
  HOOK_INPUT=$(cat package.json)
24
24
  script=$(hook_get scripts.test)
25
25
  HOOK_INPUT=$saved
26
26
  case "$script" in "" | *"no test specified"*) ;; *) printf 'npm test'; return ;; esac
27
27
  fi
28
- if [ -f composer.json ] && [ -x vendor/bin/phpunit ]; then printf 'vendor/bin/phpunit'
29
- elif { [ -f pyproject.toml ] || [ -f setup.py ]; } && command -v pytest >/dev/null 2>&1; then printf 'pytest -q'
30
- elif [ -f pubspec.yaml ] && command -v flutter >/dev/null 2>&1; then printf 'flutter test'
31
- elif [ -f go.mod ] && command -v go >/dev/null 2>&1; then printf 'go test ./...'
28
+ if [[ -f composer.json ]] && [[ -x vendor/bin/phpunit ]]; then printf 'vendor/bin/phpunit'
29
+ elif { [[ -f pyproject.toml ]] || [[ -f setup.py ]]; } && command -v pytest >/dev/null 2>&1; then printf 'pytest -q'
30
+ elif [[ -f pubspec.yaml ]] && command -v flutter >/dev/null 2>&1; then printf 'flutter test'
31
+ elif [[ -f go.mod ]] && command -v go >/dev/null 2>&1; then printf 'go test ./...'
32
32
  fi
33
33
  }
34
34
 
35
35
  cmd=$(detect_cmd)
36
- [ -n "$cmd" ] || exit 0
36
+ [[ -n "$cmd" ]] || exit 0
37
37
 
38
38
  # Empreinte de l'arbre de travail (hors journaux) : vide hors dépôt git => pas de cache.
39
39
  fingerprint=""
@@ -41,7 +41,7 @@ if git rev-parse --git-dir >/dev/null 2>&1; then
41
41
  fingerprint=$({ git status --porcelain -- . ':!.claude/logs'; git diff HEAD -- . ':!.claude/logs'; } 2>/dev/null | cksum)
42
42
  fi
43
43
  ok_file=.claude/logs/quality-gate.ok
44
- if [ -n "$fingerprint" ] && [ -f "$ok_file" ] && [ "$(cat "$ok_file")" = "$fingerprint" ]; then
44
+ if [[ -n "$fingerprint" ]] && [[ -f "$ok_file" ]] && [[ "$(cat "$ok_file")" = "$fingerprint" ]]; then
45
45
  exit 0
46
46
  fi
47
47
 
@@ -9,7 +9,7 @@ set -u
9
9
  hook_has_parser || exit 0
10
10
  hook_input
11
11
  CWD=$(hook_get cwd)
12
- [ -z "$CWD" ] || cd "$CWD" 2>/dev/null || true
12
+ [[ -z "$CWD" ]] || cd "$CWD" 2>/dev/null || true
13
13
  SESSION_ID=$(hook_get session_id)
14
14
  REASON=$(hook_get reason)
15
15
  NOW=$(date -u +%Y-%m-%dT%H:%M:%SZ)
@@ -24,13 +24,13 @@ FILES=$(git status --porcelain -- . ':!.claude' 2>/dev/null | wc -l | tr -d ' ')
24
24
 
25
25
  # Échecs d'outils du jour ; `|| true` : grep -c affiche déjà 0 et sort en 1 sans correspondance.
26
26
  FAILURES=0
27
- if [ -f .claude/logs/tool-failures.log ]; then
27
+ if [[ -f .claude/logs/tool-failures.log ]]; then
28
28
  FAILURES=$(grep -c "^\[$(date -u +%Y-%m-%d)" .claude/logs/tool-failures.log) || true
29
29
  fi
30
30
 
31
31
  STORY=none
32
32
  STATUS=unknown
33
- if [ -f .bmad/current-story.yaml ]; then
33
+ if [[ -f .bmad/current-story.yaml ]]; then
34
34
  STORY=$(grep '^id:' .bmad/current-story.yaml | head -n 1 | cut -d' ' -f2) || true
35
35
  STATUS=$(grep '^status:' .bmad/current-story.yaml | head -n 1 | cut -d' ' -f2) || true
36
36
  fi
@@ -9,12 +9,12 @@ set -u
9
9
  hook_has_parser || exit 0
10
10
  hook_input
11
11
  CWD=$(hook_get cwd)
12
- [ -z "$CWD" ] || cd "$CWD" 2>/dev/null || true
12
+ [[ -z "$CWD" ]] || cd "$CWD" 2>/dev/null || true
13
13
 
14
14
  # Variables sûres de .env.local, au format `export NAME=valeur`. Valeurs limitées à
15
15
  # [A-Za-z0-9_.:/-] : CLAUDE_ENV_FILE est sourcé par le shell, pas de métacaractères.
16
16
  # DATABASE_URL est exclue (identifiants possibles) ; commentaires et lignes vides ignorés.
17
- if [ -n "${CLAUDE_ENV_FILE:-}" ] && [ -f .env.local ]; then
17
+ if [[ -n "${CLAUDE_ENV_FILE:-}" ]] && [[ -f .env.local ]]; then
18
18
  sed -n -E 's#^(NODE_ENV|APP_ENV|APP_DEBUG|DATABASE_HOST|DATABASE_PORT|LOG_LEVEL|PORT|HOST)=([A-Za-z0-9_.:/-]*)$#export \1=\2#p' \
19
19
  .env.local >>"$CLAUDE_ENV_FILE" 2>/dev/null
20
20
  fi
@@ -29,18 +29,18 @@ json_field() {
29
29
  }
30
30
 
31
31
  CONTEXT=""
32
- add() { [ -z "$2" ] || CONTEXT="$CONTEXT$1 $2, "; }
32
+ add() { [[ -z "$2" ]] || CONTEXT="$CONTEXT$1 $2, "; }
33
33
 
34
- if [ -f composer.json ]; then
34
+ if [[ -f composer.json ]]; then
35
35
  add Symfony "$(json_field composer.json 'require.symfony/framework-bundle')"
36
36
  add PHP "$(json_field composer.json require.php)"
37
37
  fi
38
- if [ -f package.json ]; then
38
+ if [[ -f package.json ]]; then
39
39
  add React "$(json_field package.json dependencies.react)"
40
40
  add Next.js "$(json_field package.json dependencies.next)"
41
41
  fi
42
- [ ! -f pubspec.yaml ] || CONTEXT="${CONTEXT}Flutter/Dart, "
43
- [ ! -f pyproject.toml ] || CONTEXT="${CONTEXT}Python, "
42
+ [[ ! -f pubspec.yaml ]] || CONTEXT="${CONTEXT}Flutter/Dart, "
43
+ [[ ! -f pyproject.toml ]] || CONTEXT="${CONTEXT}Python, "
44
44
 
45
45
  # Services déclarés dans les fichiers Compose (noms canoniques et historiques).
46
46
  COMPOSE=$(cat compose.yaml compose.yml docker-compose.yaml docker-compose.yml 2>/dev/null)
@@ -49,5 +49,5 @@ for svc in postgres:PostgreSQL mysql:MySQL redis:Redis; do
49
49
  done
50
50
 
51
51
  CONTEXT=${CONTEXT%, }
52
- [ -z "$CONTEXT" ] || hook_context SessionStart "Project stack: $CONTEXT"
52
+ [[ -z "$CONTEXT" ]] || hook_context SessionStart "Project stack: $CONTEXT"
53
53
  exit 0
@@ -8,7 +8,8 @@
8
8
  {
9
9
  "type": "command",
10
10
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/block-dangerous-commands.sh\"",
11
- "timeout": 5
11
+ "timeout": 5,
12
+ "onFailure": "block"
12
13
  }
13
14
  ]
14
15
  }
@@ -8,7 +8,8 @@
8
8
  {
9
9
  "type": "command",
10
10
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/protect-sensitive-files.sh\"",
11
- "timeout": 5
11
+ "timeout": 5,
12
+ "onFailure": "block"
12
13
  }
13
14
  ]
14
15
  }
@@ -8,7 +8,8 @@
8
8
  {
9
9
  "type": "command",
10
10
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-network.sh\"",
11
- "timeout": 5
11
+ "timeout": 5,
12
+ "onFailure": "block"
12
13
  }
13
14
  ]
14
15
  }
@@ -7,12 +7,14 @@
7
7
  {
8
8
  "type": "command",
9
9
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-network.sh\"",
10
- "timeout": 5
10
+ "timeout": 5,
11
+ "onFailure": "block"
11
12
  },
12
13
  {
13
14
  "type": "command",
14
15
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/block-dangerous-commands.sh\"",
15
- "timeout": 5
16
+ "timeout": 5,
17
+ "onFailure": "block"
16
18
  },
17
19
  {
18
20
  "type": "command",
@@ -27,7 +29,8 @@
27
29
  {
28
30
  "type": "command",
29
31
  "command": "bash \"$CLAUDE_PROJECT_DIR/.claude/hooks/protect-sensitive-files.sh\"",
30
- "timeout": 5
32
+ "timeout": 5,
33
+ "onFailure": "block"
31
34
  }
32
35
  ]
33
36
  }
@@ -23,7 +23,7 @@ See `../../../../docs/ECOSYSTEM.md` for the full catalogue, licenses, and `.mcp.
23
23
  |------|---------|-----------|-----|
24
24
  | **caveman** | Apache-2.0 | Compress agent **output** ~65% | ✅ Integrate |
25
25
  | **code-review-graph** | MIT | AST graph → read only the blast radius on review | ✅ Integrate |
26
- | **token-savior** | MIT | Symbol index + Bash output compaction (−80%), RTK alternative | ✅ Integrate |
26
+ | **token-savior** | MIT | Symbol index + Bash output compaction, RTK alternative (−80% headline unverified; −45% cost re-measured) | 🔶 Reference |
27
27
  | **claude-token-efficient** | MIT | Anti-verbosity CLAUDE.md drop-in | ✅ Integrate |
28
28
  | **claude-hud** | MIT | Real-time context/agent/todo monitor (status line) | ✅ Integrate |
29
29
  | **context-mode** | ELv2 | Output sandboxing — license blocks commercial redistribution | 🔶 Reference |
@@ -40,6 +40,7 @@ Claude-Code-Hooks für automatisierte Qualitätskontrolle und Sicherheit, die cl
40
40
  - Die Nutzdaten kommen als **JSON über stdin**; nur der **Exit-Code 2** blockiert (jeder andere Wert ungleich null ist ein nicht blockierender Fehler).
41
41
  - Nur `hookSpecificOutput.additionalContext` erreicht das Modell; der Klartext-stdout von PreCompact wird als Compaction-Anweisung gelesen.
42
42
  - Die Skripte brauchen `jq` oder `node`, um die Nutzdaten zu lesen: Die Sicherheitswächter blockieren, wenn keines verfügbar ist, die informativen Hooks tun nichts.
43
+ - Die drei Sicherheitswächter (`guard-network`, `block-dangerous-commands`, `protect-sensitive-files`) setzen `"onFailure": "block"` (Claude Code 2.1.295+): Fehlt das Skript, stürzt es ab oder überschreitet es sein Timeout, wird der Tool-Aufruf blockiert statt ausgeführt. Blockiert ein Wächter nach einer defekten Installation jeden Aufruf, die Installation erneut ausführen oder die Hooks mit `--hooks=none` entfernen. Eine erneute Installation ergänzt `onFailure` bei Wächtern aus einer älteren Version und ersetzt nie einen selbst gesetzten Wert.
43
44
 
44
45
  ## Weitere Informationen
45
46
 
@@ -176,7 +176,7 @@ Diese Felder ermoeglichen die Optimierung von Kosten und Umfang der Sub-Agents.
176
176
 
177
177
  Claude Code kompaktiert den Kontext automatisch, wenn er sich den Fenstergrenzen naehert. Aeltere Nachrichten werden zusammengefasst, um Platz freizugeben.
178
178
 
179
- > **Append-only-Verlauf (Opus 5.5, Fable 5.x, Mythos 5.x):** Eine frühere Nachricht zu ändern, `system` oder `tools` neu aufzubauen oder alte Turns an Ort und Stelle zusammenzufassen, macht die nachfolgenden Thinking-Blöcke ungültig. Am Ende des Verlaufs anhängen statt ihn umzuschreiben. Quelle: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
179
+ > **Append-only-Verlauf (Opus 5.5, Sonnet 5.5, Haiku 5.5, Fable/Mythos 5.1 und 5.2):** Eine frühere Nachricht zu ändern, `system` oder `tools` neu aufzubauen oder alte Turns an Ort und Stelle zusammenzufassen, macht die nachfolgenden Thinking-Blöcke ungültig. Am Ende des Verlaufs anhängen statt ihn umzuschreiben. Quelle: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
180
180
 
181
181
  ### Proaktive Kompaktierung
182
182
 
@@ -365,7 +365,9 @@ Das Aufwandsniveau des Modells je nach Aufgabenkomplexität anpassen:
365
365
  > **Effort wechselt nicht das Modell:** Es steuert die Reasoning-Tiefe des aktuellen Modells; das Modell wählt man mit `/model`. Empfohlene Paare: `low` mit Haiku 5.5, `medium` mit Sonnet 5.5, `high`/`xhigh` mit Opus 5.5.
366
366
 
367
367
  > **Standard-Effort je Modell:** medium bei Opus 5.5 und Haiku 5.5; high bei Sonnet 5.5, Fable 5.2 und Legacy-Modellen.
368
- > **Fable 5.2** (`claude-fable-5-2`, $8/$40 pro Mtok) ist das aktuelle Fable-Modell, geeignet für kreative/narrative Agents; Fable 5.1 und Fable 5 bleiben verfügbar (Legacy). Bei Fable 5.2 (wie bei Mythos 5.2) ist das Denken immer aktiv (nur adaptive): Um es zu reduzieren, den Effort senken — eine Anweisung „antworte direkt“ reicht ab `medium` nicht aus. **Mythos** hat eingeschränkte Verfügbarkeit (nur Cyber Verification Program). Cache-Lesezugriffe kosten 5 % des Input-Preises bei Opus/Sonnet 5.5 und 2,5 % bei Fable 5.x und Mythos 5.x.
368
+ > **Fable 5.2** (`claude-fable-5-2`, $8/$40 pro Mtok) ist das aktuelle Fable-Modell, geeignet für kreative/narrative Agents; Fable 5.1 und Fable 5 bleiben verfügbar (Legacy). Bei Fable 5.2 (wie bei Mythos 5.2) ist das Denken immer aktiv (nur adaptive) und geht ab Effort `high` fast jeder Antwort voraus: Um es zu reduzieren, den Effort senken — eine Anweisung „antworte direkt“ reicht ab `medium` nicht aus. **Mythos** hat eingeschränkte Verfügbarkeit (nur Cyber Verification Program). Cache-Lesezugriffe kosten 5 % des Input-Preises bei Opus/Sonnet 5.5 und 2,5 % bei Fable/Mythos 5.1 und 5.2; 10 % bei den übrigen Modellen, darunter Fable 5 und Mythos 5 ([Pricing](https://platform.claude.com/docs/en/about-claude/pricing)).
369
+
370
+ > **Standard-Thinking und API-Migration:** Bei Sonnet 5.5 und Haiku 5.5 (wie bei den Legacy-Modellen Opus 5 und Sonnet 5) ist Thinking aktiv, wenn `thinking` fehlt: Content-Blöcke nach `type` statt nach Position auswählen und Spielraum in `max_tokens` lassen. Abschalten lässt es sich nur bei Effort `high` oder niedriger; bei Sonnet 5.5 ist die niedrigste Stufe `between_tools`. `budget_tokens` liefert seit Claude 4.7 einen 400-Fehler (stattdessen `effort` setzen), und das Vorbefüllen des letzten Assistant-Turns wird seit Claude 4.6 nicht mehr unterstützt. Quelle: [Prompting Best Practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
369
371
 
370
372
  ### Inaktivitaetswarnung (v2.1.84+)
371
373
 
@@ -814,8 +816,8 @@ Ergänzend zu RTK und nativen Hooks bietet das Claude-Code-Ökosystem Tools, die
814
816
  | Tool | Lizenz | Aspekt | Empf. |
815
817
  |------|--------|--------|-------|
816
818
  | **caveman** | Apache-2.0 | Komprimierung der Antworten (Output) ~65 % | ✅ Integrieren |
817
- | **code-review-graph** | MIT | AST-Graph, Blast-Radius-Lesen (−38× bis −528×) | ✅ Integrieren |
818
- | **token-savior** | MIT | Symbolindex + Bash-Komprimierung (−80 %) | ✅ Integrieren |
819
+ | **code-review-graph** | MIT | AST-Graph, Blast-Radius-Lesen (~63× Median, 6,65× vs. grep) | ✅ Integrieren |
820
+ | **token-savior** | MIT | Symbolindex + Bash-Komprimierung (−80 % nicht verifizierbar, −45 % Kosten nachgemessen) | 🔶 Referenzieren (Gewinn nicht reproduzierbar) |
819
821
  | **claude-token-efficient** | MIT | CLAUDE.md-Regeln gegen Geschwätzigkeit (~63 % Output) | ✅ Integrieren |
820
822
  | **context-mode** | ELv2 | Output-Sandbox, Kontinuität nach Kompaktierung | 🔶 Referenzieren (Lizenz) |
821
823
  | **claude-context** | MIT | Semantische Suche (Vektor-DB erforderlich) | 🔶 Referenzieren (Infra) |
@@ -248,7 +248,7 @@ Punkte |
248
248
  2. `metadata`: `sprint_id` (`sprint-{N}`), `name`, `start_date`, `end_date`, `goal`, `capacity_points`, aus `sprint-goal.md` übernommen.
249
249
  3. `stories`: ein Eintrag pro ausgewählter US: `title`, `status: ready-for-dev`, `story_points`, `epic_id`, `priority` (MoSCoW), `dependencies`, `tasks`, `history: []`.
250
250
  4. Datei für **dieselbe** `sprint_id` bereits vorhanden: bestehende Status beibehalten (eine Story nie zurücksetzen) und nur neue US ergänzen.
251
- 5. Datei für einen **anderen** Sprint vorhanden: zuerst deren Stories nach `archived_sprints.<alte sprint_id>` (`status`, `points`) verschieben, dann den neuen Sprint schreiben.
251
+ 5. Datei für einen **anderen** Sprint vorhanden: zuerst nur die **abgeschlossenen (`done`) Stories dieses vorherigen Sprints** nach `archived_sprints.<alte sprint_id>` (`status`, `points`) verschieben — nie die von `/workflow:plan` erfassten Backlog-Stories —, dann den neuen Sprint schreiben. Nicht abgeschlossene Stories bleiben in `stories` (mit aktualisierter `sprint_id`, falls dieser Schlüssel verwendet wird).
252
252
 
253
253
  ```yaml
254
254
  # .bmad/sprint-status.yaml
@@ -40,6 +40,7 @@ Claude Code hooks for automated quality control and safety, installed in your pr
40
40
  - The payload arrives as **JSON on stdin**; only **exit code 2** blocks (any other non-zero code is a non-blocking error).
41
41
  - Only `hookSpecificOutput.additionalContext` reaches the model; PreCompact plain-text stdout is read as compaction instructions.
42
42
  - Scripts need `jq` or `node` to read the payload: the safety guards block when neither is available, the informational hooks do nothing.
43
+ - The three safety guards (`guard-network`, `block-dangerous-commands`, `protect-sensitive-files`) set `"onFailure": "block"` (Claude Code 2.1.295+): if the script is missing, crashes or times out, the tool call is blocked instead of going ahead. If a guard blocks every call after a broken install, re-run the install, or remove the hooks with `--hooks=none`. Re-running the install adds `onFailure` to guards installed by an older version, and never replaces a value you set.
43
44
 
44
45
  ## More Information
45
46
 
@@ -176,7 +176,7 @@ These fields allow optimizing costs and scope of sub-agents.
176
176
 
177
177
  Claude Code automatically compacts context when approaching window limits. Older messages are summarized to free space.
178
178
 
179
- > **Append-only history (Opus 5.5, Fable 5.x, Mythos 5.x):** editing an earlier message, rebuilding `system` or `tools`, or summarizing old turns in place invalidates the thinking blocks that follow. Append to the end of the history instead of rewriting it. Source: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
179
+ > **Append-only history (Opus 5.5, Sonnet 5.5, Haiku 5.5, Fable/Mythos 5.1 and 5.2):** editing an earlier message, rebuilding `system` or `tools`, or summarizing old turns in place invalidates the thinking blocks that follow. Append to the end of the history instead of rewriting it. Source: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
180
180
 
181
181
  ### Proactive Compaction
182
182
 
@@ -365,7 +365,9 @@ Adjust the model's effort level based on task complexity:
365
365
  > **Effort does not switch models:** it sets the reasoning depth of the current model; pick the model with `/model`. Recommended pairs: `low` on Haiku 5.5, `medium` on Sonnet 5.5, `high`/`xhigh` on Opus 5.5.
366
366
 
367
367
  > **Default effort per model:** medium on Opus 5.5 and Haiku 5.5; high on Sonnet 5.5, Fable 5.2 and legacy models.
368
- > **Fable 5.2** (`claude-fable-5-2`, $8/$40 per Mtok) is the current Fable model, suitable for creative/narrative agents; Fable 5.1 and Fable 5 remain available (legacy). On Fable 5.2 (as on Mythos 5.2) thinking is always on (adaptive only): to reduce it, lower the effort — an "answer directly" instruction is not enough at `medium` and above. **Mythos** is limited availability (Cyber Verification Program only). Cache reads cost 5% of the input price on Opus/Sonnet 5.5 and 2.5% on Fable 5.x and Mythos 5.x.
368
+ > **Fable 5.2** (`claude-fable-5-2`, $8/$40 per Mtok) is the current Fable model, suitable for creative/narrative agents; Fable 5.1 and Fable 5 remain available (legacy). On Fable 5.2 (as on Mythos 5.2) thinking is always on (adaptive only) and, at `high` effort and above, precedes almost every reply: to reduce it, lower the effort — an "answer directly" instruction is not enough at `medium` and above. **Mythos** is limited availability (Cyber Verification Program only). Cache reads cost 5% of the input price on Opus/Sonnet 5.5 and 2.5% on Fable/Mythos 5.1 and 5.2; 10% on other models, including Fable 5 and Mythos 5 ([pricing](https://platform.claude.com/docs/en/about-claude/pricing)).
369
+
370
+ > **Default thinking and API migration:** on Sonnet 5.5 and Haiku 5.5 (as on the legacy Opus 5 and Sonnet 5), thinking is on when `thinking` is omitted: select content blocks by `type`, not by position, and leave room in `max_tokens`. It can only be turned off at effort `high` or lower; on Sonnet 5.5 the lowest setting is `between_tools`. `budget_tokens` returns a 400 error since Claude 4.7 (set `effort` instead), and prefilling the last assistant turn is no longer supported since Claude 4.6. Source: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
369
371
 
370
372
  ### Inactivity Alert (v2.1.84+)
371
373
 
@@ -820,8 +822,8 @@ Beyond RTK and native hooks, the Claude Code ecosystem offers tools that cover a
820
822
  | Tool | License | Angle | Rec |
821
823
  |------|---------|-------|-----|
822
824
  | **caveman** | Apache-2.0 | Output (response) compression ~65% | ✅ Integrate |
823
- | **code-review-graph** | MIT | AST graph, blast-radius reads (−38× to −528×) | ✅ Integrate |
824
- | **token-savior** | MIT | Symbol index + Bash compaction (−80%) | ✅ Integrate |
825
+ | **code-review-graph** | MIT | AST graph, blast-radius reads (~63× median, 6.65× vs grep) | ✅ Integrate |
826
+ | **token-savior** | MIT | Symbol index + Bash compaction (−80% unverified, −45% cost re-measured) | 🔶 Reference (gain not reproducible) |
825
827
  | **claude-token-efficient** | MIT | Anti-verbosity CLAUDE.md rules (~63% output) | ✅ Integrate |
826
828
  | **context-mode** | ELv2 | Output sandbox, post-compaction continuity | 🔶 Reference (license) |
827
829
  | **claude-context** | MIT | Semantic search (vector DB required) | 🔶 Reference (infra) |
@@ -248,7 +248,7 @@ Points |
248
248
  2. `metadata`: `sprint_id` (`sprint-{N}`), `name`, `start_date`, `end_date`, `goal`, `capacity_points`, taken from `sprint-goal.md`.
249
249
  3. `stories`: one entry per selected US: `title`, `status: ready-for-dev`, `story_points`, `epic_id`, `priority` (MoSCoW), `dependencies`, `tasks`, `history: []`.
250
250
  4. File already present for the **same** `sprint_id`: keep the existing statuses (never move a story back) and only add the new US.
251
- 5. File present for **another** sprint: first move its stories under `archived_sprints.<old sprint_id>` (`status`, `points`), then write the new sprint.
251
+ 5. File present for **another** sprint: first move under `archived_sprints.<old sprint_id>` (`status`, `points`) only the **finished (`done`) stories of that previous sprint** — never the backlog stories recorded by `/workflow:plan` — then write the new sprint. Unfinished stories stay in `stories` (with `sprint_id` updated if that key is used).
252
252
 
253
253
  ```yaml
254
254
  # .bmad/sprint-status.yaml
@@ -40,6 +40,7 @@ Hooks de Claude Code para el control de calidad y la seguridad automatizados, in
40
40
  - El payload llega como **JSON por stdin**; solo el **código de salida 2** bloquea (cualquier otro código distinto de cero es un error no bloqueante).
41
41
  - Solo `hookSpecificOutput.additionalContext` llega al modelo; el stdout en texto plano de PreCompact se lee como instrucciones de compactación.
42
42
  - Los scripts necesitan `jq` o `node` para leer el payload: las guardas de seguridad bloquean si no hay ninguno, los hooks informativos no hacen nada.
43
+ - Las tres guardas de seguridad (`guard-network`, `block-dangerous-commands`, `protect-sensitive-files`) declaran `"onFailure": "block"` (Claude Code 2.1.295+): si el script falta, falla o supera su timeout, la llamada a la herramienta se bloquea en lugar de continuar. Si una guarda bloquea todas las llamadas tras una instalación rota, vuelve a ejecutar la instalación o elimina los hooks con `--hooks=none`. Reinstalar añade `onFailure` a las guardas instaladas por una versión anterior, sin sustituir nunca un valor que hayas fijado.
43
44
 
44
45
  ## Más información
45
46
 
@@ -176,7 +176,7 @@ Estos campos permiten optimizar los costos y el alcance de los sub-agentes.
176
176
 
177
177
  Claude Code compacta automaticamente el contexto cuando se acerca a los limites de la ventana. Los mensajes antiguos se resumen para liberar espacio.
178
178
 
179
- > **Historial append-only (Opus 5.5, Fable 5.x, Mythos 5.x):** modificar un mensaje anterior, reconstruir `system` o `tools`, o resumir turnos antiguos en su lugar invalida los bloques de thinking siguientes. Añadir al final del historial en lugar de reescribirlo. Fuente: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
179
+ > **Historial append-only (Opus 5.5, Sonnet 5.5, Haiku 5.5, Fable/Mythos 5.1 y 5.2):** modificar un mensaje anterior, reconstruir `system` o `tools`, o resumir turnos antiguos en su lugar invalida los bloques de thinking siguientes. Añadir al final del historial en lugar de reescribirlo. Fuente: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
180
180
 
181
181
  ### Compactacion proactiva
182
182
 
@@ -365,7 +365,9 @@ Ajustar el nivel de esfuerzo del modelo según la complejidad de la tarea:
365
365
  > **El esfuerzo no cambia de modelo:** ajusta la profundidad de razonamiento del modelo actual; el modelo se elige con `/model`. Combinaciones recomendadas: `low` con Haiku 5.5, `medium` con Sonnet 5.5, `high`/`xhigh` con Opus 5.5.
366
366
 
367
367
  > **Esfuerzo por defecto según el modelo:** medium en Opus 5.5 y Haiku 5.5; high en Sonnet 5.5, Fable 5.2 y los modelos legacy.
368
- > **Fable 5.2** (`claude-fable-5-2`, $8/$40 por Mtok) es el modelo Fable actual, adecuado para agentes creativos/narrativos; Fable 5.1 y Fable 5 siguen disponibles (legacy). En Fable 5.2 (como en Mythos 5.2) el razonamiento está siempre activo (solo modo adaptive): para reducirlo, baja el esfuerzo — una instrucción «responde directamente» no basta a partir de `medium`. **Mythos** tiene disponibilidad limitada (solo Cyber Verification Program). La lectura de caché cuesta el 5 % del precio de entrada en Opus/Sonnet 5.5 y el 2,5 % en Fable 5.x y Mythos 5.x.
368
+ > **Fable 5.2** (`claude-fable-5-2`, $8/$40 por Mtok) es el modelo Fable actual, adecuado para agentes creativos/narrativos; Fable 5.1 y Fable 5 siguen disponibles (legacy). En Fable 5.2 (como en Mythos 5.2) el razonamiento está siempre activo (solo modo adaptive) y, a partir de `high`, precede a casi cada respuesta: para reducirlo, baja el esfuerzo — una instrucción «responde directamente» no basta a partir de `medium`. **Mythos** tiene disponibilidad limitada (solo Cyber Verification Program). La lectura de caché cuesta el 5 % del precio de entrada en Opus/Sonnet 5.5 y el 2,5 % en Fable/Mythos 5.1 y 5.2; 10 % en los demás modelos, incluidos Fable 5 y Mythos 5 ([pricing](https://platform.claude.com/docs/en/about-claude/pricing)).
369
+
370
+ > **Thinking por defecto y migración de la API:** en Sonnet 5.5 y Haiku 5.5 (como en los modelos legacy Opus 5 y Sonnet 5), el thinking está activo cuando se omite `thinking`: seleccionar los bloques por `type` y no por posición, y dejar margen en `max_tokens`. Solo se desactiva con esfuerzo `high` o inferior; en Sonnet 5.5 el ajuste mínimo es `between_tools`. `budget_tokens` devuelve un error 400 desde Claude 4.7 (usar `effort` en su lugar) y el prellenado del último turno del asistente ya no se admite desde Claude 4.6. Fuente: [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
369
371
 
370
372
  ### Alerta de inactividad (v2.1.84+)
371
373
 
@@ -814,8 +816,8 @@ Además de RTK y los hooks nativos, el ecosistema de Claude Code ofrece herramie
814
816
  | Herramienta | Licencia | Ángulo | Rec |
815
817
  |-------------|----------|--------|-----|
816
818
  | **caveman** | Apache-2.0 | Compresión de respuestas (output) ~65 % | ✅ Integrar |
817
- | **code-review-graph** | MIT | Grafo AST, lectura del blast radius (−38× a −528×) | ✅ Integrar |
818
- | **token-savior** | MIT | Índice simbólico + compactación de Bash (−80 %) | ✅ Integrar |
819
+ | **code-review-graph** | MIT | Grafo AST, lectura del blast radius (~63× mediana, 6,65× vs grep) | ✅ Integrar |
820
+ | **token-savior** | MIT | Índice simbólico + compactación de Bash (−80 % no verificable, −45 % de coste remedido) | 🔶 Referenciar (ganancia no reproducible) |
819
821
  | **claude-token-efficient** | MIT | Reglas CLAUDE.md anti-verbosidad (~63 % output) | ✅ Integrar |
820
822
  | **context-mode** | ELv2 | Sandbox de outputs, continuidad tras compactación | 🔶 Referenciar (licencia) |
821
823
  | **claude-context** | MIT | Búsqueda semántica (requiere base vectorial) | 🔶 Referenciar (infra) |
@@ -248,7 +248,7 @@ Puntos |
248
248
  2. `metadata`: `sprint_id` (`sprint-{N}`), `name`, `start_date`, `end_date`, `goal`, `capacity_points`, tomados de `sprint-goal.md`.
249
249
  3. `stories`: una entrada por US seleccionada: `title`, `status: ready-for-dev`, `story_points`, `epic_id`, `priority` (MoSCoW), `dependencies`, `tasks`, `history: []`.
250
250
  4. Archivo ya presente para el **mismo** `sprint_id`: conservar los estados existentes (nunca hacer retroceder una story) y añadir solo las nuevas US.
251
- 5. Archivo presente para **otro** sprint: mover primero sus stories a `archived_sprints.<sprint_id anterior>` (`status`, `points`) y luego escribir el nuevo sprint.
251
+ 5. Archivo presente para **otro** sprint: mover primero a `archived_sprints.<sprint_id anterior>` (`status`, `points`) únicamente las stories **terminadas (`done`) de ese sprint anterior** —nunca las stories de backlog registradas por `/workflow:plan`— y luego escribir el nuevo sprint. Las stories no terminadas permanecen en `stories` (con `sprint_id` actualizado si se usa esa clave).
252
252
 
253
253
  ```yaml
254
254
  # .bmad/sprint-status.yaml
@@ -40,6 +40,7 @@ Hooks Claude Code de contrôle qualité et de sécurité automatisés, installé
40
40
  - Le payload arrive en **JSON sur stdin** ; seul le **code de sortie 2** bloque (tout autre code non nul est une erreur non bloquante).
41
41
  - Seul `hookSpecificOutput.additionalContext` atteint le modèle ; le stdout en texte brut de PreCompact est lu comme consignes de compaction.
42
42
  - Les scripts ont besoin de `jq` ou `node` pour lire le payload : les gardes de sécurité bloquent si aucun n'est disponible, les hooks informatifs ne font rien.
43
+ - Les trois gardes de sécurité (`guard-network`, `block-dangerous-commands`, `protect-sensitive-files`) déclarent `"onFailure": "block"` (Claude Code 2.1.295+) : si le script manque, plante ou dépasse son timeout, l'appel d'outil est bloqué au lieu de passer. Si une garde bloque tous les appels après une installation cassée, relancez l'installation, ou retirez les hooks avec `--hooks=none`. Relancer l'installation ajoute `onFailure` aux gardes installées par une version antérieure, sans jamais remplacer une valeur que vous avez posée.
43
44
 
44
45
  ## Pour aller plus loin
45
46
 
@@ -176,7 +176,7 @@ Ces champs permettent d'optimiser les couts et le scope des sous-agents.
176
176
 
177
177
  Claude Code compacte automatiquement le contexte quand il approche les limites de la fenetre. Les messages anciens sont resumes pour liberer de l'espace.
178
178
 
179
- > **Historique append-only (Opus 5.5, Fable 5.x, Mythos 5.x) :** modifier un message antérieur, reconstruire `system` ou `tools`, ou résumer d'anciens tours sur place invalide les blocs de thinking suivants. Ajouter en fin d'historique plutôt que réécrire. Source : [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
179
+ > **Historique append-only (Opus 5.5, Sonnet 5.5, Haiku 5.5, Fable/Mythos 5.1 et 5.2) :** modifier un message antérieur, reconstruire `system` ou `tools`, ou résumer d'anciens tours sur place invalide les blocs de thinking suivants. Ajouter en fin d'historique plutôt que réécrire. Source : [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
180
180
 
181
181
  ### Compaction proactive
182
182
 
@@ -365,7 +365,9 @@ Ajuster le niveau d'effort du modele selon la complexite de la tache:
365
365
  > **L'effort ne change pas de modèle :** il règle la profondeur de raisonnement du modèle courant ; le modèle se choisit avec `/model`. Couples conseillés : `low` sur Haiku 5.5, `medium` sur Sonnet 5.5, `high`/`xhigh` sur Opus 5.5.
366
366
 
367
367
  > **Effort par défaut selon le modèle :** medium sur Opus 5.5 et Haiku 5.5 ; high sur Sonnet 5.5, Fable 5.2 et les modèles legacy.
368
- > **Fable 5.2** (`claude-fable-5-2`, 8 $/40 $ par Mtok) est le modèle Fable actuel, adapté aux agents créatifs/narratifs ; Fable 5.1 et Fable 5 restent disponibles (legacy). Sur Fable 5.2 (comme sur Mythos 5.2), la réflexion est toujours active (mode adaptive seul) : pour la réduire, baisser l'effort — une consigne « réponds directement » ne suffit pas à `medium` et au-delà. **Mythos** est en disponibilité limitée (Cyber Verification Program uniquement). La lecture de cache coûte 5 % du prix d'entrée sur Opus/Sonnet 5.5 et 2,5 % sur Fable 5.x et Mythos 5.x.
368
+ > **Fable 5.2** (`claude-fable-5-2`, 8 $/40 $ par Mtok) est le modèle Fable actuel, adapté aux agents créatifs/narratifs ; Fable 5.1 et Fable 5 restent disponibles (legacy). Sur Fable 5.2 (comme sur Mythos 5.2), la réflexion est toujours active (mode adaptive seul) et, à `high` et au-delà, précède presque chaque réponse : pour la réduire, baisser l'effort — une consigne « réponds directement » ne suffit pas à `medium` et au-delà. **Mythos** est en disponibilité limitée (Cyber Verification Program uniquement). La lecture de cache coûte 5 % du prix d'entrée sur Opus/Sonnet 5.5 et 2,5 % sur Fable/Mythos 5.1 et 5.2 ; 10 % sur les autres modèles, dont Fable 5 et Mythos 5 ([pricing](https://platform.claude.com/docs/en/about-claude/pricing)).
369
+
370
+ > **Thinking par défaut et migration API :** sur Sonnet 5.5 et Haiku 5.5 (comme les modèles *legacy* Opus 5 et Sonnet 5), le thinking est actif quand `thinking` est omis : sélectionner les blocs par `type` et non par position, et garder de la marge dans `max_tokens`. Il ne se désactive qu'à effort `high` ou moins ; sur Sonnet 5.5, le réglage minimal est `between_tools`. `budget_tokens` renvoie une erreur 400 depuis Claude 4.7 (régler l'`effort` à la place), et le préremplissage du dernier tour assistant n'est plus supporté depuis Claude 4.6. Source : [prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).
369
371
 
370
372
  ### Alerte d'inactivite (v2.1.84+)
371
373
 
@@ -784,8 +786,8 @@ En complément de RTK et des hooks natifs, l'écosystème Claude Code fournit de
784
786
  | Outil | Licence | Angle | Reco |
785
787
  |-------|---------|-------|------|
786
788
  | **caveman** | Apache-2.0 | Compression des réponses (output) ~65 % | ✅ Intégrer |
787
- | **code-review-graph** | MIT | Graphe AST, lecture du blast radius (−38× à −528×) | ✅ Intégrer |
788
- | **token-savior** | MIT | Index symbolique + compaction Bash (−80 %) | ✅ Intégrer |
789
+ | **code-review-graph** | MIT | Graphe AST, lecture du blast radius (~63× médiane, 6,65× vs grep) | ✅ Intégrer |
790
+ | **token-savior** | MIT | Index symbolique + compaction Bash (−80 % non vérifiable, −45 % de coût remesuré) | 🔶 Référencer (gain non reproductible) |
789
791
  | **claude-token-efficient** | MIT | Règles CLAUDE.md anti-verbosité (~63 % output) | ✅ Intégrer |
790
792
  | **context-mode** | ELv2 | Sandbox des outputs, continuité post-compaction | 🔶 Référencer (licence) |
791
793
  | **claude-context** | MIT | Recherche sémantique (vector DB requise) | 🔶 Référencer (infra) |