re-cinq · nextlevelshit · Apr 5, 2026 · Apr 4, 2026 · Apr 4, 2026 · Apr 4, 2026
diff --git a/.wave/pipelines/audit-closed.yaml b/.wave/pipelines/audit-closed.yaml
@@ -21,7 +21,7 @@ input:
 steps:
   - id: inventory
     persona: "{{ forge.type }}-analyst"
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -110,7 +110,7 @@ steps:
 
   - id: audit
     persona: "{{ forge.type }}-analyst"
-    model: claude-haiku
+    model: cheapest
     dependencies: [inventory]
     memory:
       inject_artifacts:
@@ -196,7 +196,7 @@ steps:
 
   - id: triage
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     dependencies: [audit]
     memory:
       inject_artifacts:

diff --git a/.wave/pipelines/audit-consolidate.yaml b/.wave/pipelines/audit-consolidate.yaml
@@ -17,7 +17,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/audit-dead-code-issue.yaml b/.wave/pipelines/audit-dead-code-issue.yaml
@@ -19,7 +19,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -80,7 +80,7 @@ steps:
 
   - id: compose-report
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     dependencies: [scan]
     memory:
       inject_artifacts:

diff --git a/.wave/pipelines/audit-dead-code-review.yaml b/.wave/pipelines/audit-dead-code-review.yaml
@@ -19,7 +19,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -85,7 +85,7 @@ steps:
 
   - id: compose
     persona: summarizer
-    model: claude-haiku
+    model: cheapest
     dependencies: [scan]
     memory:
       inject_artifacts:

diff --git a/.wave/pipelines/audit-dead-code-scan.yaml b/.wave/pipelines/audit-dead-code-scan.yaml
@@ -12,7 +12,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     max_concurrent_agents: 3
     contexts: [configuration]
     workspace:

diff --git a/.wave/pipelines/audit-dead-code.yaml b/.wave/pipelines/audit-dead-code.yaml
@@ -19,7 +19,7 @@ steps:
   - id: scan
     persona: navigator
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/audit-doc-scan.yaml b/.wave/pipelines/audit-doc-scan.yaml
@@ -12,7 +12,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     max_concurrent_agents: 3
     contexts: [delivery, quality]
     workspace:

diff --git a/.wave/pipelines/audit-doc.yaml b/.wave/pipelines/audit-doc.yaml
@@ -18,7 +18,7 @@ steps:
   - id: scan-changes
     persona: navigator
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       type: worktree
       branch: "{{ pipeline_id }}"

diff --git a/.wave/pipelines/audit-dual.yaml b/.wave/pipelines/audit-dual.yaml
@@ -31,7 +31,7 @@ steps:
   # ── Track A: Code Quality ──────────────────────────────────────────
   - id: quality-scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -67,7 +67,7 @@ steps:
 
   - id: quality-detail
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     dependencies: [quality-scan]
     memory:
       strategy: fresh
@@ -104,7 +104,7 @@ steps:
   # ── Track B: Security ──────────────────────────────────────────────
   - id: security-scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -130,7 +130,7 @@ steps:
 
   - id: security-detail
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     dependencies: [security-scan]
     memory:
       strategy: fresh
@@ -167,7 +167,7 @@ steps:
   # ── Merge: Converge both tracks ────────────────────────────────────
   - id: merge
     persona: summarizer
-    model: claude-haiku
+    model: cheapest
     dependencies: [quality-detail, security-detail]
     memory:
       strategy: fresh

diff --git a/.wave/pipelines/audit-duplicates.yaml b/.wave/pipelines/audit-duplicates.yaml
@@ -12,7 +12,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     max_concurrent_agents: 3
     contexts: [configuration]
     workspace:

diff --git a/.wave/pipelines/audit-dx.yaml b/.wave/pipelines/audit-dx.yaml
@@ -19,7 +19,7 @@ steps:
     persona: navigator
     contexts: [quality]
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/audit-junk-code.yaml b/.wave/pipelines/audit-junk-code.yaml
@@ -18,7 +18,7 @@ steps:
   - id: scan
     persona: navigator
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/audit-security.yaml b/.wave/pipelines/audit-security.yaml
@@ -27,7 +27,7 @@ steps:
   - id: scan
     persona: navigator
     contexts: [security]
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/audit-unwired.yaml b/.wave/pipelines/audit-unwired.yaml
@@ -12,7 +12,7 @@ input:
 steps:
   - id: scan
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     max_concurrent_agents: 3
     contexts: [execution, configuration]
     workspace:

diff --git a/.wave/pipelines/audit-ux.yaml b/.wave/pipelines/audit-ux.yaml
@@ -19,7 +19,7 @@ steps:
     persona: navigator
     contexts: [quality]
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/doc-changelog.yaml b/.wave/pipelines/doc-changelog.yaml
@@ -14,7 +14,7 @@ input:
 steps:
   - id: analyze-commits
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       type: worktree
       branch: "{{ pipeline_id }}"

diff --git a/.wave/pipelines/doc-explain.yaml b/.wave/pipelines/doc-explain.yaml
@@ -14,7 +14,7 @@ input:
 steps:
   - id: explore
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./
@@ -62,7 +62,7 @@ steps:
 
   - id: analyze
     persona: planner
-    model: claude-haiku
+    model: cheapest
     dependencies: [explore]
     memory:
       inject_artifacts:

diff --git a/.wave/pipelines/doc-fix.yaml b/.wave/pipelines/doc-fix.yaml
@@ -14,7 +14,7 @@ input:
 steps:
   - id: scan-changes
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/doc-onboard.yaml b/.wave/pipelines/doc-onboard.yaml
@@ -14,7 +14,7 @@ input:
 steps:
   - id: survey
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/impl-feature.yaml b/.wave/pipelines/impl-feature.yaml
@@ -28,7 +28,7 @@ steps:
   - id: explore
     persona: navigator
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/impl-hotfix.yaml b/.wave/pipelines/impl-hotfix.yaml
@@ -36,7 +36,7 @@ hooks:
 steps:
   - id: investigate
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/impl-improve.yaml b/.wave/pipelines/impl-improve.yaml
@@ -28,7 +28,7 @@ steps:
   - id: assess
     persona: navigator
     max_concurrent_agents: 3
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/impl-issue-core.yaml b/.wave/pipelines/impl-issue-core.yaml
@@ -35,7 +35,7 @@ steps:
   - id: fetch-assess
     persona: implementer
     contexts: [configuration]
-    model: claude-haiku
+    model: cheapest
     workspace:
       type: worktree
       branch: "{{ pipeline_id }}"

diff --git a/.wave/pipelines/impl-issue.yaml b/.wave/pipelines/impl-issue.yaml
@@ -50,7 +50,7 @@ steps:
   - id: fetch-assess
     persona: implementer
     contexts: [configuration]
-    model: claude-haiku
+    model: cheapest
     workspace:
       type: worktree
       branch: "{{ pipeline_id }}"
@@ -160,7 +160,7 @@ steps:
         - type: agent_review
           persona: navigator
           criteria_path: .wave/contracts/impl-review-criteria.md
-          model: claude-haiku-4-5-20251001
+          model: cheapest
           token_budget: 8000
           timeout: 90s
           context:

diff --git a/.wave/pipelines/impl-prototype.yaml b/.wave/pipelines/impl-prototype.yaml
@@ -293,7 +293,7 @@ steps:
 
   - id: pr-review
     persona: auditor
-    model: claude-haiku
+    model: cheapest
     dependencies: [pr-create]
 
     exec:

diff --git a/.wave/pipelines/impl-recinq.yaml b/.wave/pipelines/impl-recinq.yaml
@@ -75,7 +75,7 @@ steps:
   # ── Diamond 1: Discover (DIVERGENT) ──────────────────────────────────
   - id: diverge
     persona: provocateur
-    model: claude-haiku
+    model: cheapest
     dependencies: [gather]
     memory:
       inject_artifacts:

diff --git a/.wave/pipelines/impl-refactor.yaml b/.wave/pipelines/impl-refactor.yaml
@@ -15,7 +15,7 @@ input:
 steps:
   - id: analyze
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     workspace:
       mount:
         - source: ./

diff --git a/.wave/pipelines/impl-review-loop.yaml b/.wave/pipelines/impl-review-loop.yaml
@@ -46,7 +46,7 @@ steps:
 
   - id: rework
     type: pipeline
-    pipeline: ops-pr-rework
+    pipeline: ops-pr-fix-review
     dependencies: [check-verdict]
     input: "{{ implement.output.pr_url }}"
 

diff --git a/.wave/pipelines/impl-smart-route.yaml b/.wave/pipelines/impl-smart-route.yaml
@@ -12,7 +12,7 @@ input:
 steps:
   - id: assess
     persona: navigator
-    model: claude-haiku
+    model: cheapest
     exec:
       type: prompt
       source: |

diff --git a/.wave/pipelines/impl-speckit.yaml b/.wave/pipelines/impl-speckit.yaml
@@ -31,7 +31,7 @@ hooks:
 steps:
   - id: specify
     persona: implementer
-    model: claude-haiku
+    model: cheapest
     workspace:
       type: worktree
       branch: "{{ pipeline_id }}"
@@ -56,7 +56,7 @@ steps:
 
   - id: clarify
     persona: implementer
-    model: claude-haiku
+    model: cheapest
     dependencies: [specify]
     memory:
       inject_artifacts:
@@ -173,7 +173,7 @@ steps:
 
   - id: analyze
     persona: implementer
-    model: claude-haiku
+    model: cheapest
     dependencies: [checklist]
     memory:
       inject_artifacts:
@@ -228,7 +228,7 @@ steps:
         - type: agent_review
           persona: navigator
           criteria_path: .wave/contracts/impl-review-criteria.md
-          model: claude-haiku-4-5-20251001
+          model: cheapest
           token_budget: 8000
           timeout: 90s
           context:

diff --git a/.wave/pipelines/ops-debug.yaml b/.wave/pipelines/ops-debug.yaml
@@ -25,7 +25,7 @@ chat_context:
 steps:
   - id: reproduce
     persona: debugger
-    model: claude-haiku
+    model: cheapest
     thread: debug
     workspace:
       mount: