cc-devflow 4.5.7 → 4.5.9

package/docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json

@@ -6,63 +6,30 @@
   "requirementId": "REQ-002",
   "requirementVersion": "REQ-002.v2",
   "planningMeta": {
-    "reqPlanSkillVersion": "3.8.1",
+    "reqPlanSkillVersion": "3.8.5",
     "designVersion": "design.v2",
     "approvedAt": null,
-    "approvedBy": null,
     "basedOnOption": "Option B",
-    "requirementBrief": {
-      "problemStatement": "Admins onboarding larger teams need predictable bulk invite behavior for duplicates, invalid rows, and seat limits.",
-      "solutionSummary": "Freeze deterministic CSV row outcomes before executing the bulk invite flow.",
-      "actors": [
-        "workspace admin",
-        "support operator"
-      ],
-      "userStories": [
-        {
-          "id": "US-001",
-          "actor": "workspace admin",
-          "want": "upload a CSV of invite emails",
-          "benefit": "invite many collaborators without one-by-one entry",
-          "acceptance": [
-            "Mixed valid rows produce visible accepted outcomes"
-          ]
-        },
-        {
-          "id": "US-002",
-          "actor": "workspace admin",
-          "want": "see duplicate, invalid, and over-limit row states",
-          "benefit": "understand what happened without support help",
-          "acceptance": [
-            "Every skipped or rejected row has a reason"
-          ]
-        }
-      ],
-      "edgeOrRecoveryStories": [
-        {
-          "id": "US-EDGE-001",
-          "actor": "workspace admin",
-          "boundary": "CSV contains duplicates, invalid emails, or over-limit rows",
-          "desiredOutcome": "safe rows can proceed while bad rows are explained",
-          "acceptance": [
-            "Rule matrix covers duplicate, invalid, and seat-limit cases"
-          ]
-        }
-      ],
-      "implementationDecisions": [
-        "Freeze one row-outcome matrix before execution resumes"
-      ],
-      "testingDecisions": [
-        "Test row semantics through bulk-import rules and the admin upload flow"
-      ],
-      "outOfScope": [
-        "enterprise SCIM provisioning",
-        "background job redesign",
-        "rollback wizard for partial success"
-      ],
-      "furtherNotes": [
-        "Design remains blocked until duplicate and seat-limit semantics are approved"
-      ]
+    "aiLeverageDecisionLens": {
+      "realUserOrOperator": "workspace admin onboarding larger teams",
+      "statusQuoWorkaround": "paste invite emails one by one or track them in spreadsheets",
+      "humanTeamEffortForFullScope": "multiple weeks across product, engineering, billing, and support review",
+      "ccAgentEffortForFullScope": "several hours once row semantics are approved, but unbounded before that",
+      "aiCompressionRatio": "high after semantics freeze, low while the product contract is ambiguous",
+      "completeLakeBoundary": "row classification, admin upload flow, billing-seat checks, and audit mapping after rule approval",
+      "oceanBoundary": "SCIM, background retries, rollback wizard, and unspecified billing semantics",
+      "scopeRecommendation": "sharp-wedge",
+      "costModel": {
+        "agentTime": "medium after rule approval",
+        "humanReviewTime": "high until semantics are approved",
+        "verificationCost": "bulk rules, admin flow, billing, and audit tests",
+        "maintenanceCost": "medium after semantics freeze",
+        "failureCost": "high if billing or audit outcomes drift",
+        "reversibility": "limited once execution semantics ship"
+      },
+      "verdict": "sharp-wedge",
+      "missingEvidenceOrPivotReason": "none at plan approval; cc-check later reopened duplicate, invalid-row, partial-success, and seat-limit semantics before retry or cc-act",
+      "impactOnApprovedDirection": "allow bounded import-path execution, then block final proof if review finds row semantics drift"
     },
     "externalBestPractice": {
       "needed": true,
@@ -172,31 +139,9 @@
         "impact": "cc-do retry and cc-act must stay blocked until this is answered",
         "status": "asked"
       }
-    ],
-    "aiLeverageDecisionLens": {
-      "realUserOrOperator": "workspace admin onboarding larger teams",
-      "statusQuoWorkaround": "paste invite emails one by one or track them in spreadsheets",
-      "humanTeamEffortForFullScope": "multiple weeks across product, engineering, billing, and support review",
-      "ccAgentEffortForFullScope": "several hours once row semantics are approved, but unbounded before that",
-      "aiCompressionRatio": "high after semantics freeze, low while the product contract is ambiguous",
-      "completeLakeBoundary": "row classification, admin upload flow, billing-seat checks, and audit mapping after rule approval",
-      "oceanBoundary": "SCIM, background retries, rollback wizard, and unspecified billing semantics",
-      "scopeRecommendation": "sharp-wedge",
-      "costModel": {
-        "agentTime": "medium after rule approval",
-        "humanReviewTime": "high until semantics are approved",
-        "verificationCost": "bulk rules, admin flow, billing, and audit tests",
-        "maintenanceCost": "medium after semantics freeze",
-        "failureCost": "high if billing or audit outcomes drift",
-        "reversibility": "limited once execution semantics ship"
-      },
-      "verdict": "sharp-wedge",
-      "missingEvidenceOrPivotReason": "none at plan approval; cc-check later reopened duplicate, invalid-row, partial-success, and seat-limit semantics before retry or cc-act",
-      "impactOnApprovedDirection": "allow bounded import-path execution, then block final proof if review finds row semantics drift"
-    }
+    ]
   },
   "currentTaskId": null,
-  "activePhase": null,
   "tasks": [
     {
       "id": "T001",
@@ -245,7 +190,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "red",
+      "verticalSlice": "Slice 1",
+      "testSeam": {
+        "entry": "bulk invite rules and admin upload UI behavior",
+        "behaviorAsserted": "dangerous row outcomes are reproduced as explicit failing tests",
+        "specStyleTestName": "dangerous row outcomes are reproduced as explicit failing tests",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the bulk invite rule and admin panel tests through their public flows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": "Fails before the behavior exists"
+      }
     },
     {
       "id": "T002",
@@ -297,7 +257,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "green",
+      "verticalSlice": "Slice 2",
+      "testSeam": {
+        "entry": "bulk invite rules and admin upload UI behavior",
+        "behaviorAsserted": "bulk invite rows receive deterministic initial states",
+        "specStyleTestName": "bulk invite rows receive deterministic initial states",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the bulk invite rule and admin panel tests through their public flows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": ""
+      }
     },
     {
       "id": "T003",
@@ -348,7 +323,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "red",
+      "verticalSlice": "Slice 3",
+      "testSeam": {
+        "entry": "bulk invite rules and admin upload UI behavior",
+        "behaviorAsserted": "mixed row outcomes are visible as UI failures before implementation",
+        "specStyleTestName": "mixed row outcomes are visible as UI failures before implementation",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the bulk invite rule and admin panel tests through their public flows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": "Fails before the behavior exists"
+      }
     },
     {
       "id": "T004",
@@ -400,7 +390,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "green",
+      "verticalSlice": "Slice 3",
+      "testSeam": {
+        "entry": "bulk invite rules and admin upload UI behavior",
+        "behaviorAsserted": "admins can see accepted, skipped, and rejected rows in the panel",
+        "specStyleTestName": "admins can see accepted, skipped, and rejected rows in the panel",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the bulk invite rule and admin panel tests through their public flows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": ""
+      }
     },
     {
       "id": "T005",
@@ -456,7 +461,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "evidence",
+      "verticalSlice": "Slice 3",
+      "testSeam": {
+        "entry": "bulk invite rules and admin upload UI behavior",
+        "behaviorAsserted": "fresh verification evidence exists for cc-check",
+        "specStyleTestName": "fresh verification evidence exists for cc-check",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the bulk invite rule and admin panel tests through their public flows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": ""
+      }
     }
   ],
   "metadata": {

package/docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md

@@ -4,7 +4,7 @@
 
 - Requirement version: `REQ-002.v2`
 - Design version: `design.v2`
-- CC-Plan skill version: `3.8.1`
+- CC-Plan skill version: `3.8.5`
 - Source roadmap item: `RM-010`
 - Source roadmap version: `roadmap.v2`
 
@@ -37,50 +37,94 @@
   - audit row-result contract
 - Parallel boundaries: no parallel execution until row-outcome semantics are frozen
 
+## Execution Protocol
+
+ClaudeCode / Codex 执行本计划时，必须把本文件当成任务模板合同，而不是普通 TODO 列表。
+
+- Template source: `assets/TASKS_TEMPLATE.md`
+- Task selection: read `planning/task-manifest.json.currentTaskId`; if empty, run the ready-task selector before choosing work.
+- Task block rule: read the full task block before coding; title-only execution is invalid.
+- Completion rule: after verification and review gates pass, run the completion script; do not manually edit checkbox, status, or `currentTaskId`.
+- Completion failure: if the script fails, fix the missing checkpoint / review / dependency evidence and rerun it. Do not bypass it by editing JSON or Markdown.
+
+```bash
+SCRIPT_ROOT=".claude/skills/cc-do/scripts"
+if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then
+  SCRIPT_ROOT=".codex/skills/cc-do/scripts"
+fi
+bash "$SCRIPT_ROOT/select-ready-tasks.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json
+bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task <task-id>
+```
+
 ## Phase 1: Rule Matrix
 
 - [x] T001 [TEST] Add failing tests for duplicate and over-limit row outcomes (dependsOn:none) `src/invite/bulk-import.test.ts`
   Goal: 先把最危险的 bulk invite 行为变成红灯。
+  TDD phase: red
   Files: `src/invite/bulk-import.test.ts`
   Read first: `design.md`, `src/invite/bulk-import.ts`
   Verification: `npm test -- src/invite/bulk-import.test.ts`
   Evidence: failing output
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task T001`; do not hand-edit status.
+  Test seam: bulk invite rules and admin upload UI behavior
+  Public verification path: Run the bulk invite rule and admin panel tests through their public flows
+  Allowed mocks: file upload boundary / billing / seat limit boundary
   Ready when: 当前规则文件已经可定位
 
 - [x] T002 [IMPL] Implement the initial bulk invite row classification (dependsOn:T001) `src/invite/bulk-import.ts`
   Goal: 给 duplicate / over-limit 行为一个最小实现。
+  TDD phase: green
   Files: `src/invite/bulk-import.ts`
   Read first: `design.md`, `src/invite/bulk-import.test.ts`
   Verification: `npm test -- src/invite/bulk-import.test.ts`
   Evidence: passing output + checkpoint
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task T002`; do not hand-edit status.
+  Test seam: bulk invite rules and admin upload UI behavior
+  Public verification path: Run the bulk invite rule and admin panel tests through their public flows
+  Allowed mocks: file upload boundary / billing / seat limit boundary
   Ready when: T001 已见红
 
 ## Phase 2: Surface Integration
 
 - [x] T003 [TEST] Add failing admin panel tests for mixed CSV results (dependsOn:T002) `src/admin/BulkInvitePanel.test.tsx`
   Goal: 在 UI 层看到 mixed row outcomes 的真实表现。
+  TDD phase: red
   Files: `src/admin/BulkInvitePanel.test.tsx`
   Read first: `design.md`, `src/admin/BulkInvitePanel.tsx`
   Verification: `npm test -- src/admin/BulkInvitePanel.test.tsx`
   Evidence: failing output
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task T003`; do not hand-edit status.
+  Test seam: bulk invite rules and admin upload UI behavior
+  Public verification path: Run the bulk invite rule and admin panel tests through their public flows
+  Allowed mocks: file upload boundary / billing / seat limit boundary
   Ready when: T002 已完成
 
 - [x] T004 [IMPL] Render bulk invite results in the admin panel (dependsOn:T003) `src/admin/BulkInvitePanel.tsx`
   Goal: 把 row outcomes 接进管理界面。
+  TDD phase: green
   Files: `src/admin/BulkInvitePanel.tsx`
   Read first: `design.md`, `src/admin/BulkInvitePanel.test.tsx`
   Verification: `npm test -- src/admin/BulkInvitePanel.test.tsx`
   Evidence: passing output + review notes
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task T004`; do not hand-edit status.
+  Test seam: bulk invite rules and admin upload UI behavior
+  Public verification path: Run the bulk invite rule and admin panel tests through their public flows
+  Allowed mocks: file upload boundary / billing / seat limit boundary
   Ready when: T003 已见红
 
 ## Phase 3: Verification
 
 - [x] T005 Run mixed-surface checks and gather review evidence (dependsOn:T004) `bulk invite quality gates`
   Goal: 为 `cc-check` 准备 bulk invite 的新鲜证据。
+  TDD phase: evidence
   Files: `src/admin/BulkInvitePanel.tsx`, `src/invite/bulk-import.ts`
   Read first: `tasks.md`, `task-manifest.json`
   Verification:
   - `npm test -- src/invite/bulk-import.test.ts`
   - `npm test -- src/admin/BulkInvitePanel.test.tsx`
   Evidence: passing output + review notes
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/task-manifest.json --tasks docs/examples/full-design-blocked/changes/REQ-002-bulk-invite-import/planning/tasks.md --task T005`; do not hand-edit status.
+  Test seam: bulk invite rules and admin upload UI behavior
+  Public verification path: Run the bulk invite rule and admin panel tests through their public flows
+  Allowed mocks: file upload boundary / billing / seat limit boundary
   Ready when: T004 已完成

package/docs/examples/full-design-blocked/roadmap.json

@@ -5,7 +5,7 @@
   },
   "meta": {
     "roadmapVersion": "roadmap.v2",
-    "skillVersion": "5.2.0",
+    "skillVersion": "5.3.0",
     "status": "active",
     "lastUpdated": "2026-04-19",
     "currentFocusStage": "Stage 2"

package/docs/examples/local-handoff/BACKLOG.md

@@ -5,7 +5,7 @@
 ## Backlog Meta
 
 - Roadmap version: `roadmap.v2`
-- Skill version: `5.2.0`
+- Skill version: `5.3.0`
 - Last synced: `2026-04-19`
 - Current focus stage: `Stage 2`
 - Roadmap state source: `roadmap.json`

package/docs/examples/local-handoff/README.md

@@ -4,7 +4,7 @@
 
 - Example version: `1.0.0`
 - Last reviewed: `2026-04-17`
-- Bound skills: `cc-roadmap@5.2.0`, `cc-plan@3.8.1`, `cc-do@1.6.2`, `cc-check@1.10.1`, `cc-act@1.8.2`
+- Bound skills: `cc-roadmap@5.3.0`, `cc-plan@3.8.5`, `cc-do@1.6.5`, `cc-check@1.10.1`, `cc-act@1.8.7`
 
 This example shows verified work that is **ready to move forward**, but `cc-act` still chooses `local-handoff`.
 

package/docs/examples/local-handoff/ROADMAP.md

@@ -3,7 +3,7 @@
 ## Roadmap Meta
 
 - Roadmap version: `roadmap.v3`
-- Skill version: `5.2.0`
+- Skill version: `5.3.0`
 - Status: `active`
 - Last updated: `2026-04-16`
 - Owner / decider: `product-owner`

package/docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/design.md

@@ -4,7 +4,7 @@
 
 - Requirement version: `REQ-003.v1`
 - Design version: `design.v1`
-- CC-Plan skill version: `3.8.1`
+- CC-Plan skill version: `3.8.5`
 - Requirement ID: `REQ-003`
 - Design mode: `tiny-design`
 - Why this stays `tiny-design`: the patch adds one export action inside the existing admin audit UI without changing data contracts

package/docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json

@@ -6,44 +6,30 @@
   "requirementId": "REQ-003",
   "requirementVersion": "REQ-003.v1",
   "planningMeta": {
-    "reqPlanSkillVersion": "3.8.1",
+    "reqPlanSkillVersion": "3.8.5",
     "designVersion": "design.v1",
     "approvedAt": "2026-04-16T13:10:00.000Z",
-    "approvedBy": "user",
     "basedOnOption": "Tiny design card",
-    "requirementBrief": {
-      "problemStatement": "Admins can review audit summary rows in the UI, but weekly reporting requires manual copying.",
-      "solutionSummary": "Add a CSV download action for the currently visible audit summary rows.",
-      "actors": [
-        "workspace admin reviewing weekly activity"
-      ],
-      "userStories": [
-        {
-          "id": "US-001",
-          "actor": "workspace admin",
-          "want": "download visible audit summary rows as CSV",
-          "benefit": "include them in weekly reporting without manual copying",
-          "acceptance": [
-            "Panel behavior test proves the export action uses current summary rows"
-          ]
-        }
-      ],
-      "edgeOrRecoveryStories": [],
-      "implementationDecisions": [
-        "Export only rows already visible in the panel",
-        "Keep CSV as the only output format"
-      ],
-      "testingDecisions": [
-        "Test through the admin panel action and visible row data"
-      ],
-      "outOfScope": [
-        "JSON export",
-        "scheduled reporting",
-        "shared reporting backend"
-      ],
-      "furtherNotes": [
-        "Richer machine-readable exports should become a separate requirement"
-      ]
+    "aiLeverageDecisionLens": {
+      "realUserOrOperator": "workspace admin preparing weekly review notes",
+      "statusQuoWorkaround": "manual copy of visible audit rows",
+      "humanTeamEffortForFullScope": "about one day for an engineer to implement, test, and document the local export",
+      "ccAgentEffortForFullScope": "about 30 minutes for visible-row CSV export plus targeted test and lint",
+      "aiCompressionRatio": "roughly 10x for the bounded local export path",
+      "completeLakeBoundary": "visible-row CSV export, panel action, current data source, targeted panel test, and lint",
+      "oceanBoundary": "JSON export, scheduled reporting, shared reporting backend, and cross-panel reporting platform",
+      "scopeRecommendation": "boil-lake",
+      "costModel": {
+        "agentTime": "low",
+        "humanReviewTime": "low",
+        "verificationCost": "targeted panel test plus lint",
+        "maintenanceCost": "low while scoped to visible rows",
+        "failureCost": "admins keep copying rows manually",
+        "reversibility": "reversible UI action"
+      },
+      "verdict": "boil-lake",
+      "missingEvidenceOrPivotReason": "",
+      "impactOnApprovedDirection": "complete the local visible-row export lake while deferring reporting platform work"
     },
     "externalBestPractice": {
       "needed": false,
@@ -89,31 +75,9 @@
         "impact": "cc-do exports only visible rows from the current panel and avoids new reporting contracts",
         "status": "answered"
       }
-    ],
-    "aiLeverageDecisionLens": {
-      "realUserOrOperator": "workspace admin preparing weekly review notes",
-      "statusQuoWorkaround": "manual copy of visible audit rows",
-      "humanTeamEffortForFullScope": "about one day for an engineer to implement, test, and document the local export",
-      "ccAgentEffortForFullScope": "about 30 minutes for visible-row CSV export plus targeted test and lint",
-      "aiCompressionRatio": "roughly 10x for the bounded local export path",
-      "completeLakeBoundary": "visible-row CSV export, panel action, current data source, targeted panel test, and lint",
-      "oceanBoundary": "JSON export, scheduled reporting, shared reporting backend, and cross-panel reporting platform",
-      "scopeRecommendation": "boil-lake",
-      "costModel": {
-        "agentTime": "low",
-        "humanReviewTime": "low",
-        "verificationCost": "targeted panel test plus lint",
-        "maintenanceCost": "low while scoped to visible rows",
-        "failureCost": "admins keep copying rows manually",
-        "reversibility": "reversible UI action"
-      },
-      "verdict": "boil-lake",
-      "missingEvidenceOrPivotReason": "",
-      "impactOnApprovedDirection": "complete the local visible-row export lake while deferring reporting platform work"
-    }
+    ]
   },
   "currentTaskId": null,
-  "activePhase": null,
   "tasks": [
     {
       "id": "T001",
@@ -163,7 +127,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "red",
+      "verticalSlice": "Slice 1",
+      "testSeam": {
+        "entry": "admin audit panel UI behavior",
+        "behaviorAsserted": "the missing download action is reproduced as a failing test",
+        "specStyleTestName": "the missing download action is reproduced as a failing test",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the audit summary panel test and observe CSV export through visible rows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": "Fails before the behavior exists"
+      }
     },
     {
       "id": "T002",
@@ -215,7 +194,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "green",
+      "verticalSlice": "Slice 2",
+      "testSeam": {
+        "entry": "admin audit panel UI behavior",
+        "behaviorAsserted": "admins can trigger a CSV download from the existing summary panel",
+        "specStyleTestName": "admins can trigger a CSV download from the existing summary panel",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the audit summary panel test and observe CSV export through visible rows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": ""
+      }
     },
     {
       "id": "T003",
@@ -271,7 +265,22 @@
       },
       "status": "passed",
       "attempts": 1,
-      "maxRetries": 1
+      "maxRetries": 1,
+      "tddPhase": "evidence",
+      "verticalSlice": "Slice 3",
+      "testSeam": {
+        "entry": "admin audit panel UI behavior",
+        "behaviorAsserted": "fresh verification evidence exists for local handoff",
+        "specStyleTestName": "fresh verification evidence exists for local handoff",
+        "oneLogicalBehavior": true,
+        "publicVerificationPath": "Run the audit summary panel test and observe CSV export through visible rows",
+        "implementationDetailRisk": "low"
+      },
+      "feedbackLoop": {
+        "type": "automated-test",
+        "determinism": "deterministic",
+        "expectedFailure": ""
+      }
     }
   ],
   "metadata": {

package/docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/tasks.md

@@ -4,7 +4,7 @@
 
 - Requirement version: `REQ-003.v1`
 - Design version: `design.v1`
-- CC-Plan skill version: `3.8.1`
+- CC-Plan skill version: `3.8.5`
 - Source roadmap item: `RM-020`
 - Source roadmap version: `roadmap.v3`
 
@@ -36,32 +36,66 @@
   - panel placement
 - Parallel boundaries: none
 
+## Execution Protocol
+
+ClaudeCode / Codex 执行本计划时，必须把本文件当成任务模板合同，而不是普通 TODO 列表。
+
+- Template source: `assets/TASKS_TEMPLATE.md`
+- Task selection: read `planning/task-manifest.json.currentTaskId`; if empty, run the ready-task selector before choosing work.
+- Task block rule: read the full task block before coding; title-only execution is invalid.
+- Completion rule: after verification and review gates pass, run the completion script; do not manually edit checkbox, status, or `currentTaskId`.
+- Completion failure: if the script fails, fix the missing checkpoint / review / dependency evidence and rerun it. Do not bypass it by editing JSON or Markdown.
+
+```bash
+SCRIPT_ROOT=".claude/skills/cc-do/scripts"
+if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then
+  SCRIPT_ROOT=".codex/skills/cc-do/scripts"
+fi
+bash "$SCRIPT_ROOT/select-ready-tasks.sh" --manifest docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json
+bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json --tasks docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/tasks.md --task <task-id>
+```
+
 ## Phase 1: Foundation
 
 - [x] T001 [TEST] Add a failing test for the missing download action (dependsOn:none) `src/admin/AuditSummaryPanel.test.tsx`
   Goal: 证明现在还没有可下载的 summary export。
+  TDD phase: red
   Files: `src/admin/AuditSummaryPanel.test.tsx`
   Read first: `design.md`, `src/admin/AuditSummaryPanel.tsx`
   Verification: `npm test -- src/admin/AuditSummaryPanel.test.tsx`
   Evidence: failing output
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json --tasks docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/tasks.md --task T001`; do not hand-edit status.
+  Test seam: admin audit panel UI behavior
+  Public verification path: Run the audit summary panel test and observe CSV export through visible rows
+  Allowed mocks: download / blob boundary
   Ready when: 当前 audit summary test 已可运行
 
 - [x] T002 [IMPL] Add the download summary action (dependsOn:T001) `src/admin/AuditSummaryPanel.tsx`
   Goal: 用最小实现让 summary export 真的可用。
+  TDD phase: green
   Files: `src/admin/AuditSummaryPanel.tsx`
   Read first: `design.md`, `src/admin/AuditSummaryPanel.test.tsx`
   Verification: `npm test -- src/admin/AuditSummaryPanel.test.tsx`
   Evidence: passing output + checkpoint
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json --tasks docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/tasks.md --task T002`; do not hand-edit status.
+  Test seam: admin audit panel UI behavior
+  Public verification path: Run the audit summary panel test and observe CSV export through visible rows
+  Allowed mocks: download / blob boundary
   Ready when: T001 已见红
 
 ## Phase 2: Verify
 
 - [x] T003 Run targeted checks and collect handoff evidence (dependsOn:T002) `audit summary quality gates`
   Goal: 为 `cc-check` 和 `cc-act` 留下这次导出改动的真实证据。
+  TDD phase: evidence
   Files: `src/admin/AuditSummaryPanel.tsx`, `src/admin/AuditSummaryPanel.test.tsx`
   Read first: `tasks.md`, `task-manifest.json`
   Verification:
   - `npm test -- src/admin/AuditSummaryPanel.test.tsx`
   - `npm run lint -- src/admin/AuditSummaryPanel.tsx`
   Evidence: passing output + clean lint output
+  Completion: after verification evidence and required checkpoint/review records exist, run `SCRIPT_ROOT=".claude/skills/cc-do/scripts"; if [[ ! -d "$SCRIPT_ROOT" && -d ".codex/skills/cc-do/scripts" ]]; then SCRIPT_ROOT=".codex/skills/cc-do/scripts"; fi; bash "$SCRIPT_ROOT/mark-task-complete.sh" --manifest docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/task-manifest.json --tasks docs/examples/local-handoff/changes/REQ-003-audit-log-export/planning/tasks.md --task T003`; do not hand-edit status.
+  Test seam: admin audit panel UI behavior
+  Public verification path: Run the audit summary panel test and observe CSV export through visible rows
+  Allowed mocks: download / blob boundary
   Ready when: T002 已完成