takt 0.38.0 → 0.40.0

package/README.md

@@ -86,7 +86,7 @@ takt run
 ### Manage results
 
 ```bash
-# List task branches — merge, retry, force-fail, or delete
+# List task branches — merge, retry, requeue, force-fail, or delete
 takt list
 ```
 
@@ -150,7 +150,7 @@ See the [Builtin Catalog](./docs/builtin-catalog.md) for all workflows and perso
 |---------|-------------|
 | `takt` | Talk to AI, refine requirements, execute or queue tasks |
 | `takt run` | Execute all pending tasks |
-| `takt list` | Manage task branches (merge, retry, force-fail, instruct, delete) |
+| `takt list` | Manage task branches (merge, retry, requeue, force-fail, instruct, delete) |
 | `takt #N` | Execute GitHub Issue as task |
 | `takt eject` | Copy builtin workflows/facets for customization |
 | `takt workflow init` | Create a new workflow scaffold |

package/builtins/en/facets/instructions/arbitrate.md

@@ -1,14 +1,14 @@
-The ai_review (reviewer) and ai_fix (coder) disagree.
+The ai-antipattern-review-1st (reviewer) and ai-antipattern-fix (coder) disagree.
 
-- ai_review flagged issues and issued a REJECT
-- ai_fix reviewed and determined "no fix needed"
+- ai-antipattern-review-1st flagged issues and issued a REJECT
+- ai-antipattern-fix reviewed and determined "no fix needed"
 
 Review both outputs and arbitrate which judgment is valid.
 
 **Reports to reference:**
-- AI review results: {report:ai-review.md}
+- AI review results: {report:ai-antipattern-review.md}
 
 **Judgment criteria:**
-- Whether ai_review's findings are specific and point to real issues in the code
-- Whether ai_fix's rebuttal has evidence (file verification results, test results)
+- Whether ai-antipattern-review-1st's findings are specific and point to real issues in the code
+- Whether ai-antipattern-fix's rebuttal has evidence (file verification results, test results)
 - Whether the findings are non-blocking (record only) level or actually require fixes

package/builtins/en/facets/instructions/{loop-monitor-ai-fix.md → loop-monitor-ai-antipattern-fix.md}

@@ -1,10 +1,10 @@
-The ai_review ↔ ai_fix loop has repeated {cycle_count} times.
+The ai-antipattern-review-1st ↔ ai-antipattern-fix loop has repeated {cycle_count} times.
 
 Review the reports from each cycle and determine whether this loop
 is healthy (making progress) or unproductive (repeating the same issues).
 
 **Reports to reference:**
-- AI Review results: {report:ai-review.md}
+- AI Review results: {report:ai-antipattern-review.md}
 
 **Judgment criteria:**
 - Are the same finding_ids persisting across multiple cycles?

package/builtins/en/facets/instructions/review-arch.md

@@ -1,5 +1,5 @@
 Focus on reviewing **architecture and design**.
-Do not review AI-specific issues (already covered by the ai_review step).
+Do not review AI-specific issues (already covered by the ai-antipattern-review-1st step).
 
 **Review criteria:**
 - Structural and design validity

package/builtins/en/facets/instructions/review-cqrs-es.md

@@ -1,5 +1,5 @@
 Review the changes from the perspective of CQRS (Command Query Responsibility Segregation) and Event Sourcing.
-AI-specific issue review is not needed (already covered by the ai_review step).
+AI-specific issue review is not needed (already covered by the ai-antipattern-review-1st step).
 
 **Review criteria:**
 - Aggregate design validity

package/builtins/en/facets/instructions/review-terraform.md

@@ -1,5 +1,5 @@
 Focus on reviewing **Terraform convention compliance**.
-Do not review AI-specific issues (already covered by the ai_review step).
+Do not review AI-specific issues (already covered by the ai-antipattern-review-1st step).
 
 **Review criteria:**
 - Variable declaration compliance (type, description, sensitive)

package/builtins/en/facets/policies/review.md

@@ -150,6 +150,25 @@ Test file length and duplication are warning-level maintainability concerns by d
   - inability to detect regressions
 - "Too long" or "duplicated" alone is not sufficient for `REJECT`
 
+## Handling Changelog and History Files
+
+Files or sections that record point-in-time facts (e.g., `CHANGELOG.md`, `RELEASE_NOTES.md`, `MIGRATION.md`) are history, not specifications of the current code. Judge them by their correctness as history.
+
+| Target | Judgment |
+|--------|----------|
+| Past entry's config keys, API names, or behaviors do not match current code | REJECT prohibited |
+| Records that were correct at the time of the relevant release | Modification requests prohibited |
+| Factual errors in newly added entries (relative to the target release) | REJECT allowed |
+| Markdown formatting issues, duplication, broken links, obvious typos | REJECT or Warning allowed |
+
+### Judgment Criteria
+
+- History records "what changed at that point in time," not "how the system currently works"
+- Even if names or behaviors have been changed in current code, that is not grounds to rewrite past entries
+- To request modification of a past entry, demonstrate that it was incorrect even at the relevant release point
+- Identify history files/sections by file name (`CHANGELOG.md`, etc.) or conventional headings (`### Changed`, `### Added`, dated release headings)
+- Do not REJECT a history file or section based solely on disagreement with current schema or current config keys
+
 ## Boy Scout Rule
 
 Leave it better than you found it.

package/builtins/en/workflow-categories.yaml

@@ -2,12 +2,14 @@ workflow_categories:
   🚀 Quick Start:
     workflows:
       - default
+      - default-mini
       - default-high
       - frontend
       - backend
       - dual
   ⚡ Mini:
     workflows:
+      - default-mini
       - frontend-mini
       - backend-mini
       - backend-cqrs-mini

package/builtins/en/workflows/auto-improvement-loop.yaml

@@ -6,6 +6,7 @@ max_steps: infinite
 initial_step: route_context
 schemas:
   followup-task: followup-task
+  pr-followup-task: pr-followup-task
 steps:
   - name: route_context
     mode: system
@@ -37,6 +38,10 @@ steps:
       - type: issue_selection
         source: current_project
         as: selected_issue
+      - type: issue_list
+        source: current_project
+        as: tracked_issues
+        exclude_selected_from: selected_issue
     rules:
       - when: context.route_context.selected_pr.exists == true
         next: plan_from_existing_pr
@@ -62,22 +67,39 @@ steps:
       - task_exists: {context:route_context.task.exists}
       - branch_exists: {context:route_context.branch.exists}
 
-      Do not implement anything. Output only the next task instruction in task_markdown.
-      Output the next action in action.
-      If needed, output the issue creation plan in issue.
       Current execution context:
       - task_exists: {context:route_context.task.exists}
       - branch_exists: {context:route_context.branch.exists}
 
+      Open managed PR snapshot for overlap checks:
+      {context:route_context.prs}
+
+      Other open issue metadata for overlap checks:
+      {context:route_context.tracked_issues}
+      - other_open_issue_count: {context:route_context.tracked_issues.length}
+
+      Prioritize user value. Prefer bug fix > missing capability completion > UX/quality improvement > cosmetic.
+      Plan exactly one task.
+      Only propose a task when it has a concrete deliverable and explicit completion criteria.
+      Avoid cosmetic-only changes unless they clearly unlock meaningful user value.
+      Treat the current issue itself as in-scope, not as a duplicate.
+      Avoid tasks that materially duplicate or overlap with an open PR or the other open issues listed above.
+
+      Do not implement anything. Output only the next task instruction in task_markdown.
+      Output the next action in action.
+      Always output task_markdown and issue.
+      For wait_before_next_scan, set task_markdown to an empty string and issue to { create: false, labels: [] }.
+      For an existing issue task, set issue.create to false unless a new GitHub Issue is explicitly needed.
+
       Allowed action values:
       - enqueue_new_task
-      - noop
+      - wait_before_next_scan
     structured_output:
       schema_ref: followup-task
     rules:
       - when: structured.plan_from_issue.action == "enqueue_new_task"
         next: enqueue_from_issue
-      - when: structured.plan_from_issue.action == "noop"
+      - when: structured.plan_from_issue.action == "wait_before_next_scan"
         next: wait_before_next_scan
       - when: "true"
         next: ABORT
@@ -112,19 +134,35 @@ steps:
       Inspect the current codebase,
       and plan exactly one next improvement task.
 
+      Open managed PR snapshot for overlap checks:
+      {context:route_context.prs}
+
+      Open issue metadata already tracked:
+      {context:route_context.tracked_issues}
+      - open_issue_count: {context:route_context.tracked_issues.length}
+
+      Prioritize user value. Prefer bug fix > missing capability completion > UX/quality improvement > cosmetic.
+      Plan exactly one task.
+      Only propose a task when it has a concrete deliverable and explicit completion criteria.
+      Avoid cosmetic-only changes unless they clearly unlock meaningful user value.
+      Avoid tasks that materially duplicate or overlap with an open PR or tracked open issue.
+      If you cannot identify a clearly distinct, high-value task, choose wait_before_next_scan.
+
       Do not implement anything. Output only the next task instruction in task_markdown.
       Output the next action in action.
-      If needed, output the issue creation plan in issue.
+      Always output task_markdown and issue.
+      For wait_before_next_scan, set task_markdown to an empty string and issue to { create: false, labels: [] }.
+      For a fresh improvement task, set issue.create to true so the task is backed by a GitHub Issue.
 
       Allowed action values:
       - enqueue_new_task
-      - noop
+      - wait_before_next_scan
     structured_output:
       schema_ref: followup-task
     rules:
       - when: structured.plan_fresh_improvement.action == "enqueue_new_task"
         next: enqueue_fresh
-      - when: structured.plan_fresh_improvement.action == "noop"
+      - when: structured.plan_fresh_improvement.action == "wait_before_next_scan"
         next: wait_before_next_scan
       - when: "true"
         next: ABORT
@@ -165,36 +203,33 @@ steps:
       - head_branch: {context:route_context.selected_pr.head_branch}
 
       Output the next action in action.
-      If a follow-up implementation task is needed, output it in task_markdown.
-      If content should be left on the PR, output it in pr_comment_markdown.
+      Always output task_markdown.
+      For prepare_merge or reject_pr, set task_markdown to an empty string.
+      For enqueue_from_pr, output the follow-up implementation task in task_markdown.
 
       Allowed action values:
-      - comment_on_pr
       - enqueue_from_pr
       - prepare_merge
-      - noop
+      - reject_pr
     structured_output:
-      schema_ref: followup-task
+      schema_ref: pr-followup-task
     rules:
-      - when: structured.plan_from_existing_pr.action == "comment_on_pr"
-        next: comment_on_existing_pr
       - when: structured.plan_from_existing_pr.action == "enqueue_from_pr"
         next: enqueue_from_pr
       - when: structured.plan_from_existing_pr.action == "prepare_merge"
         next: prepare_merge
-      - when: structured.plan_from_existing_pr.action == "noop"
-        next: wait_before_next_scan
+      - when: structured.plan_from_existing_pr.action == "reject_pr"
+        next: reject_pr
       - when: "true"
         next: ABORT
 
-  - name: comment_on_existing_pr
+  - name: reject_pr
     mode: system
     effects:
-      - type: comment_pr
+      - type: close_pr
         pr: "{context:route_context.selected_pr.number}"
-        body: "{structured:plan_from_existing_pr.pr_comment_markdown}"
     rules:
-      - when: effect.comment_on_existing_pr.comment_pr.success == true
+      - when: effect.reject_pr.close_pr.success == true
         next: wait_before_next_scan
       - when: "true"
         next: ABORT

package/builtins/en/workflows/backend-cqrs-mini.yaml

@@ -92,7 +92,7 @@ steps:
           format: coder-decisions
   - name: reviewers
     parallel:
-      - name: ai_review
+      - name: ai-antipattern-review-2nd
         edit: false
         persona: ai-antipattern-reviewer
         policy:
@@ -106,14 +106,14 @@ steps:
               - Grep
               - WebSearch
               - WebFetch
-        instruction: review-ai
+        instruction: ai-antipattern-review
         rules:
           - condition: No AI-specific issues
           - condition: AI-specific issues found
         output_contracts:
           report:
-            - name: ai-review.md
-              format: ai-review
+            - name: ai-antipattern-review.md
+              format: ai-antipattern-review
       - name: supervise
         edit: false
         persona: supervisor
@@ -149,12 +149,12 @@ steps:
       - condition: all("AI-specific issues found", "Requirements unmet, tests failing")
         next: fix_both
       - condition: any("AI-specific issues found")
-        next: ai_fix
+        next: ai-antipattern-fix
       - condition: any("Requirements unmet, tests failing")
         next: supervise_fix
   - name: fix_both
     parallel:
-      - name: ai_fix_parallel
+      - name: ai-antipattern-fix-parallel
         edit: true
         persona: coder
         policy:
@@ -180,7 +180,7 @@ steps:
           - condition: AI Reviewer's issues fixed
           - condition: No fix needed (verified target files/spec)
           - condition: Cannot proceed, insufficient info
-        instruction: ai-fix
+        instruction: ai-antipattern-fix
       - name: supervise_fix_parallel
         edit: true
         persona: coder
@@ -212,7 +212,7 @@ steps:
         next: reviewers
       - condition: any("No fix needed (verified target files/spec)", "Cannot proceed, insufficient info")
         next: implement
-  - name: ai_fix
+  - name: ai-antipattern-fix
     edit: true
     persona: coder
     policy:
@@ -243,7 +243,7 @@ steps:
         next: implement
       - condition: Cannot proceed, insufficient info
         next: implement
-    instruction: ai-fix
+    instruction: ai-antipattern-fix
   - name: supervise_fix
     edit: true
     persona: coder

package/builtins/en/workflows/backend-cqrs.yaml

@@ -10,15 +10,15 @@ max_steps: 60
 initial_step: plan
 loop_monitors:
   - cycle:
-      - ai_review
-      - ai_fix
+      - ai-antipattern-review-1st
+      - ai-antipattern-fix
     threshold: 3
     judge:
       persona: supervisor
-      instruction: loop-monitor-ai-fix
+      instruction: loop-monitor-ai-antipattern-fix
       rules:
         - condition: Healthy (making progress)
-          next: ai_review
+          next: ai-antipattern-review-1st
         - condition: Unproductive (no improvement)
           next: reviewers_1
   - cycle:
@@ -126,11 +126,11 @@ steps:
     instruction: implement
     rules:
       - condition: Implementation is complete
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: No implementation (report only)
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: Cannot proceed with implementation
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: User input required
         next: implement
         requires_user_input: true
@@ -141,7 +141,7 @@ steps:
           format: coder-scope
         - name: coder-decisions.md
           format: coder-decisions
-  - name: ai_review
+  - name: ai-antipattern-review-1st
     edit: false
     persona: ai-antipattern-reviewer
     policy:
@@ -155,17 +155,17 @@ steps:
           - Grep
           - WebSearch
           - WebFetch
-    instruction: ai-review
+    instruction: ai-antipattern-review
     rules:
       - condition: No AI-specific issues found
         next: reviewers_1
       - condition: AI-specific issues detected
-        next: ai_fix
+        next: ai-antipattern-fix
     output_contracts:
       report:
-        - name: ai-review.md
-          format: ai-review
-  - name: ai_fix
+        - name: ai-antipattern-review.md
+          format: ai-antipattern-review
+  - name: ai-antipattern-fix
     edit: true
     persona: coder
     policy:
@@ -188,16 +188,16 @@ steps:
           - Bash
           - WebSearch
           - WebFetch
-    instruction: ai-fix
+    instruction: ai-antipattern-fix
     pass_previous_response: false
     rules:
       - condition: AI Reviewer's issues have been fixed
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: No fix needed (verified target files/spec)
-        next: ai_no_fix
+        next: ai-antipattern-no-fix
       - condition: Unable to proceed with fixes
-        next: ai_no_fix
-  - name: ai_no_fix
+        next: ai-antipattern-no-fix
+  - name: ai-antipattern-no-fix
     edit: false
     persona: architecture-reviewer
     policy: review
@@ -208,9 +208,9 @@ steps:
           - Glob
           - Grep
     rules:
-      - condition: ai_review's findings are valid (fix required)
-        next: ai_fix
-      - condition: ai_fix's judgment is valid (no fix needed)
+      - condition: ai-antipattern-review-1st's findings are valid (fix required)
+        next: ai-antipattern-fix
+      - condition: ai-antipattern-fix's judgment is valid (no fix needed)
         next: reviewers_1
     instruction: arbitrate
   - name: reviewers_1
@@ -263,6 +263,28 @@ steps:
           report:
             - name: testing-review.md
               format: testing-review
+      - name: ai-antipattern-review-2nd
+        edit: false
+        persona: ai-antipattern-reviewer
+        policy:
+          - review
+          - ai-antipattern
+        provider_options:
+          claude:
+            allowed_tools:
+              - Read
+              - Glob
+              - Grep
+              - WebSearch
+              - WebFetch
+        instruction: ai-antipattern-review
+        rules:
+          - condition: approved
+          - condition: needs_fix
+        output_contracts:
+          report:
+            - name: ai-antipattern-review.md
+              format: ai-antipattern-review
     rules:
       - condition: all("approved")
         next: reviewers_2

package/builtins/en/workflows/backend-mini.yaml

@@ -90,7 +90,7 @@ steps:
           format: coder-decisions
   - name: reviewers
     parallel:
-      - name: ai_review
+      - name: ai-antipattern-review-2nd
         edit: false
         persona: ai-antipattern-reviewer
         policy:
@@ -104,14 +104,14 @@ steps:
               - Grep
               - WebSearch
               - WebFetch
-        instruction: review-ai
+        instruction: ai-antipattern-review
         rules:
           - condition: No AI-specific issues
           - condition: AI-specific issues found
         output_contracts:
           report:
-            - name: ai-review.md
-              format: ai-review
+            - name: ai-antipattern-review.md
+              format: ai-antipattern-review
       - name: supervise
         edit: false
         persona: supervisor
@@ -146,12 +146,12 @@ steps:
       - condition: all("AI-specific issues found", "Requirements unmet, tests failing")
         next: fix_both
       - condition: any("AI-specific issues found")
-        next: ai_fix
+        next: ai-antipattern-fix
       - condition: any("Requirements unmet, tests failing")
         next: supervise_fix
   - name: fix_both
     parallel:
-      - name: ai_fix_parallel
+      - name: ai-antipattern-fix-parallel
         edit: true
         persona: coder
         policy:
@@ -176,7 +176,7 @@ steps:
           - condition: AI Reviewer's issues fixed
           - condition: No fix needed (verified target files/spec)
           - condition: Cannot proceed, insufficient info
-        instruction: ai-fix
+        instruction: ai-antipattern-fix
       - name: supervise_fix_parallel
         edit: true
         persona: coder
@@ -207,7 +207,7 @@ steps:
         next: reviewers
       - condition: any("No fix needed (verified target files/spec)", "Cannot proceed, insufficient info")
         next: implement
-  - name: ai_fix
+  - name: ai-antipattern-fix
     edit: true
     persona: coder
     policy:
@@ -237,7 +237,7 @@ steps:
         next: implement
       - condition: Cannot proceed, insufficient info
         next: implement
-    instruction: ai-fix
+    instruction: ai-antipattern-fix
   - name: supervise_fix
     edit: true
     persona: coder

package/builtins/en/workflows/backend.yaml

@@ -10,15 +10,15 @@ max_steps: 60
 initial_step: plan
 loop_monitors:
   - cycle:
-      - ai_review
-      - ai_fix
+      - ai-antipattern-review-1st
+      - ai-antipattern-fix
     threshold: 3
     judge:
       persona: supervisor
-      instruction: loop-monitor-ai-fix
+      instruction: loop-monitor-ai-antipattern-fix
       rules:
         - condition: Healthy (making progress)
-          next: ai_review
+          next: ai-antipattern-review-1st
         - condition: Unproductive (no improvement)
           next: reviewers_1
   - cycle:
@@ -123,11 +123,11 @@ steps:
     instruction: implement
     rules:
       - condition: Implementation is complete
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: No implementation (report only)
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: Cannot proceed with implementation
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: User input required
         next: implement
         requires_user_input: true
@@ -138,7 +138,7 @@ steps:
           format: coder-scope
         - name: coder-decisions.md
           format: coder-decisions
-  - name: ai_review
+  - name: ai-antipattern-review-1st
     edit: false
     persona: ai-antipattern-reviewer
     policy:
@@ -152,17 +152,17 @@ steps:
           - Grep
           - WebSearch
           - WebFetch
-    instruction: ai-review
+    instruction: ai-antipattern-review
     rules:
       - condition: No AI-specific issues found
         next: reviewers_1
       - condition: AI-specific issues detected
-        next: ai_fix
+        next: ai-antipattern-fix
     output_contracts:
       report:
-        - name: ai-review.md
-          format: ai-review
-  - name: ai_fix
+        - name: ai-antipattern-review.md
+          format: ai-antipattern-review
+  - name: ai-antipattern-fix
     edit: true
     persona: coder
     policy:
@@ -184,16 +184,16 @@ steps:
           - Bash
           - WebSearch
           - WebFetch
-    instruction: ai-fix
+    instruction: ai-antipattern-fix
     pass_previous_response: false
     rules:
       - condition: AI Reviewer's issues have been fixed
-        next: ai_review
+        next: ai-antipattern-review-1st
       - condition: No fix needed (verified target files/spec)
-        next: ai_no_fix
+        next: ai-antipattern-no-fix
       - condition: Unable to proceed with fixes
-        next: ai_no_fix
-  - name: ai_no_fix
+        next: ai-antipattern-no-fix
+  - name: ai-antipattern-no-fix
     edit: false
     persona: architecture-reviewer
     policy: review
@@ -204,9 +204,9 @@ steps:
           - Glob
           - Grep
     rules:
-      - condition: ai_review's findings are valid (fix required)
-        next: ai_fix
-      - condition: ai_fix's judgment is valid (no fix needed)
+      - condition: ai-antipattern-review-1st's findings are valid (fix required)
+        next: ai-antipattern-fix
+      - condition: ai-antipattern-fix's judgment is valid (no fix needed)
         next: reviewers_1
     instruction: arbitrate
   - name: reviewers_1
@@ -259,6 +259,28 @@ steps:
           report:
             - name: testing-review.md
               format: testing-review
+      - name: ai-antipattern-review-2nd
+        edit: false
+        persona: ai-antipattern-reviewer
+        policy:
+          - review
+          - ai-antipattern
+        provider_options:
+          claude:
+            allowed_tools:
+              - Read
+              - Glob
+              - Grep
+              - WebSearch
+              - WebFetch
+        instruction: ai-antipattern-review
+        rules:
+          - condition: approved
+          - condition: needs_fix
+        output_contracts:
+          report:
+            - name: ai-antipattern-review.md
+              format: ai-antipattern-review
     rules:
       - condition: all("approved")
         next: reviewers_2