prizmkit 1.1.140 → 1.1.142

package/bundled/dev-pipeline/tests/test_generate_bugfix_prompt.py

@@ -472,7 +472,7 @@ class TestBugfixCheckpointGeneration:
         existing["bugfix_state"] = bugfix_checkpoint_metadata(
             ".prizmkit/bugfix/B-001",
             stage="test",
-            status="TEST_BLOCKED",
+            status="failed",
             stage_result="TEST_BLOCKED",
             repair_round=2,
             next_stage="prizmkit-test",
@@ -481,6 +481,7 @@ class TestBugfixCheckpointGeneration:
 
         merged = merge_bugfix_checkpoint_state(existing, fresh, str(tmp_path))
 
+        assert merged["bugfix_state"]["status"] == "failed"
         assert merged["bugfix_state"]["stage_result"] == "TEST_BLOCKED"
         assert merged["bugfix_state"]["resume_from"] == "prizmkit-test"
 
@@ -500,66 +501,44 @@ class TestBugfixCheckpointGeneration:
         test_step = checkpoint["steps"][4]
         assert test_step["depends_on"] == "S04"
         assert test_step["required_artifacts"] == [
-            ".prizmkit/test/evidence/*/manifest.json",
-            ".prizmkit/test/evidence/*/verdict.json",
-            ".prizmkit/test/evidence/*/validation.json",
+            ".prizmkit/bugfix/B-001/test-report.md",
+            ".prizmkit/bugfix/B-001/test-result.json",
         ]
         assert checkpoint["steps"][5]["depends_on"] == "S05"
 
-    def test_bugfix_repair_routing_preserves_review_and_test_boundaries(self):
+    def test_bugfix_result_routing_stops_test_non_pass(self):
         review = route_bugfix_repair("REVIEW_NEEDS_FIXES", repair_round=0)
-        test_infra = route_bugfix_repair(
-            "TEST_FAIL", "test-infrastructure", repair_round=1,
-        )
-        production = route_bugfix_repair("TEST_FAIL", "public-interface", repair_round=2)
+        needs_fixes = route_bugfix_repair("TEST_NEEDS_FIXES", repair_round=1)
+        blocked = route_bugfix_repair("TEST_BLOCKED", repair_round=1)
 
         assert review["route"] == [
             "prizmkit-implement", "prizmkit-code-review", "prizmkit-test",
         ]
-        assert test_infra["route"] == ["prizmkit-implement", "prizmkit-test"]
-        assert production["route"] == [
-            "prizmkit-implement", "prizmkit-code-review", "prizmkit-test",
-        ]
-        assert production["repair_round"] == 3
-
-    def test_bugfix_repair_routing_blocks_unsafe_blocked_and_exhausted_results(self):
-        unknown = route_bugfix_repair("TEST_FAIL", "unknown", repair_round=1)
-        blocked = route_bugfix_repair("TEST_BLOCKED", "production", repair_round=1)
-        exhausted = route_bugfix_repair("TEST_FAIL", "production", repair_round=3)
-
-        assert unknown["blocked"] is True
-        assert unknown["route"] == []
+        assert review["repair_round"] == 1
+        assert needs_fixes["blocked"] is True
+        assert needs_fixes["route"] == []
+        assert needs_fixes["blocked_reason"] == "test_needs_fixes"
         assert blocked["blocked"] is True
         assert blocked["resumable"] is False
         assert blocked["next_stage"] is None
         assert blocked["terminal_status"] == "WORKFLOW_BLOCKED"
         assert blocked["blocked_reason"] == "test_blocked"
-        assert blocked["route"] == []
-        assert "cannot safely edit production" in blocked["reason"]
-        assert exhausted["blocked"] is True
-        assert exhausted["repair_round"] == 3
 
-    def test_bugfix_checkpoint_metadata_preserves_artifact_and_evidence_handoff(self):
+    def test_bugfix_checkpoint_metadata_has_no_evidence_handoff(self):
         metadata = bugfix_checkpoint_metadata(
             ".prizmkit/bugfix/B-001",
-            evidence={
-                "manifest_path": ".prizmkit/test/evidence/id/manifest.json",
-                "verdict_path": ".prizmkit/test/evidence/id/verdict.json",
-                "validation_path": ".prizmkit/test/evidence/id/validation.json",
-            },
             stage="test",
-            status="TEST_FAIL",
-            stage_result="TEST_FAIL",
-            repair_scope="test-infrastructure",
-            repair_round=2,
+            status="failed",
+            stage_result="TEST_NEEDS_FIXES",
+            repair_round=0,
+            next_stage=None,
+            resume_from=None,
         )
 
         assert metadata["artifact_dir"] == ".prizmkit/bugfix/B-001"
-        assert metadata["authoritative_evidence_paths"]["validation"].endswith(
-            "/validation.json"
-        )
-        assert metadata["repair_round"] == 2
-        assert metadata["resume_from"] == "prizmkit-implement"
+        assert metadata["status"] == "failed"
+        assert metadata["stage_result"] == "TEST_NEEDS_FIXES"
+        assert "authoritative_evidence_paths" not in metadata
 
     def test_manual_bugfix_uses_automatic_verification_and_keeps_commit_gate(self):
         sections = assemble_bugfix_sections(

package/bundled/dev-pipeline/tests/test_generate_refactor_prompt.py

@@ -59,12 +59,12 @@ class TestRefactorCheckpointMerge:
         existing["refactor_state"] = refactor_checkpoint_metadata(
             ".prizmkit/refactor/R-008",
             stage="test",
-            status="TEST_FAIL",
-            stage_result="TEST_FAIL",
+            status="failed",
+            stage_result="TEST_NEEDS_FIXES",
             repair_scope="public-interface",
             repair_round=2,
-            next_stage="prizmkit-implement",
-            resume_from="prizmkit-implement",
+            next_stage=None,
+            resume_from=None,
         )
         existing["steps"][0]["status"] = "skipped"
 
@@ -72,6 +72,7 @@ class TestRefactorCheckpointMerge:
 
         assert merged["steps"][0]["status"] == "skipped"
         assert merged["steps"][1]["status"] == "pending"
+        assert merged["refactor_state"]["status"] == "failed"
         assert merged["refactor_state"]["repair_scope"] == "public-interface"
         assert merged["refactor_state"]["repair_round"] == 2
 
@@ -215,7 +216,7 @@ def test_template_source_is_clean_for_headless_prompt_contract():
     assert "/prizmkit-test scope=this-change artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/" in content
     assert "local_commit_authorized=true" in content
     assert "push_authorized" not in content
-    assert "Behavior-preservation checks remain supporting evidence" in content
+    assert "Behavior-preservation context remains supporting input" in content
 
 
 # ---------------------------------------------------------------------------
@@ -349,44 +350,43 @@ class TestRefactorCheckpointGeneration:
             "completion-summary",
         ]
         assert checkpoint["steps"][4]["depends_on"] == "S04"
+        assert checkpoint["steps"][4]["required_artifacts"] == [
+            ".prizmkit/refactor/R-001/test-report.md",
+            ".prizmkit/refactor/R-001/test-result.json",
+        ]
         assert checkpoint["steps"][5]["depends_on"] == "S05"
 
-    def test_refactor_repair_routing_preserves_behavior_review_and_test(self):
-        test_infra = route_refactor_repair(
-            "TEST_FAIL", "test-infrastructure", repair_round=1,
-        )
-        production = route_refactor_repair("TEST_FAIL", "public-interface", repair_round=1)
-        blocked = route_refactor_repair("TEST_BLOCKED", "production", repair_round=1)
+    def test_refactor_result_routing_stops_test_non_pass(self):
+        review = route_refactor_repair("REVIEW_NEEDS_FIXES", repair_round=0)
+        needs_fixes = route_refactor_repair("TEST_NEEDS_FIXES", repair_round=1)
+        blocked = route_refactor_repair("TEST_BLOCKED", repair_round=1)
 
-        assert test_infra["route"] == ["prizmkit-implement", "prizmkit-test"]
-        assert production["route"] == [
+        assert review["route"] == [
             "prizmkit-implement", "prizmkit-code-review", "prizmkit-test",
         ]
-        unknown = route_refactor_repair("TEST_FAIL", "contract", repair_round=1)
-        exhausted = route_refactor_repair("TEST_FAIL", "production", repair_round=3)
-        assert unknown["blocked"] is True
-        assert unknown["route"] == []
-        assert exhausted["blocked"] is True
-        assert exhausted["repair_round"] == 3
+        assert needs_fixes["blocked"] is True
+        assert needs_fixes["route"] == []
+        assert needs_fixes["blocked_reason"] == "test_needs_fixes"
         assert blocked["blocked"] is True
         assert blocked["route"] == []
         assert blocked["next_stage"] is None
         assert blocked["terminal_status"] == "WORKFLOW_BLOCKED"
         assert blocked["blocked_reason"] == "test_blocked"
 
-    def test_refactor_checkpoint_metadata_preserves_artifact_and_evidence_handoff(self):
+    def test_refactor_checkpoint_metadata_has_no_evidence_handoff(self):
         metadata = refactor_checkpoint_metadata(
             ".prizmkit/refactor/R-001",
-            evidence={"manifest_path": "manifest.json"},
             stage="test",
-            status="TEST_PASS",
+            status="completed",
             stage_result="TEST_PASS",
             next_stage="prizmkit-retrospective",
             resume_from="prizmkit-retrospective",
         )
 
         assert metadata["artifact_dir"] == ".prizmkit/refactor/R-001"
-        assert metadata["authoritative_evidence_paths"]["manifest"] == "manifest.json"
+        assert metadata["status"] == "completed"
+        assert metadata["stage_result"] == "TEST_PASS"
+        assert "authoritative_evidence_paths" not in metadata
         assert metadata["next_stage"] == "prizmkit-retrospective"
         assert metadata["resume_from"] == "prizmkit-retrospective"
 
@@ -397,7 +397,7 @@ class TestRefactorCheckpointGeneration:
             "/prizmkit-test scope=this-change artifact_dir=.prizmkit/refactor/R-001/"
         ) < prompt.index("### Headless Commit Authorization")
         assert "prizmkit-workflow" not in prompt
-        assert "Behavior-preservation checks collected during planning and implementation remain supporting evidence" in prompt
+        assert "Existing behavior-preservation context is input to testing" in prompt
         assert '"local_commit_authorized": true' in prompt
         assert "push_authorized" not in prompt
         assert "Remote publication is controlled by the runtime" in prompt

package/bundled/dev-pipeline/tests/test_python_runner_parity.py

@@ -92,12 +92,6 @@ def _item_id_from_next_result(kind, result):
 
 
 def _write_semantic_completion_checkpoint(project_root, artifact_dir, workflow_type="feature-pipeline"):
-    validator = (
-        project_root / ".claude" / "command-assets" / "prizmkit-test"
-        / "scripts" / "validate_test_evidence.py"
-    )
-    validator.parent.mkdir(parents=True, exist_ok=True)
-    validator.write_text("import sys\nsys.exit(0)\n", encoding="utf-8")
     contracts = {
         "feature-pipeline": (
             "feature_state",
@@ -123,25 +117,6 @@ def _write_semantic_completion_checkpoint(project_root, artifact_dir, workflow_t
         ),
     }
     state_key, skills = contracts[workflow_type]
-    evidence_id = "semantic-evidence"
-    evidence_dir = project_root / ".prizmkit" / "test" / "evidence" / evidence_id
-    evidence_dir.mkdir(parents=True, exist_ok=True)
-    evidence = {
-        "manifest": evidence_dir / "manifest.json",
-        "verdict": evidence_dir / "verdict.json",
-        "validation": evidence_dir / "validation.json",
-    }
-    evidence["manifest"].write_text(
-        json.dumps({"evidence_id": evidence_id, "final_verdict": "TEST_PASS"}),
-        encoding="utf-8",
-    )
-    evidence["verdict"].write_text(json.dumps({"verdict": "TEST_PASS"}), encoding="utf-8")
-    evidence["validation"].write_text(
-        json.dumps({"verdict": "TEST_PASS", "result": "passed"}), encoding="utf-8"
-    )
-    relative_evidence = {
-        key: str(path.relative_to(project_root)) for key, path in evidence.items()
-    }
     steps = []
     previous = None
     for index, skill in enumerate(skills, start=1):
@@ -157,10 +132,38 @@ def _write_semantic_completion_checkpoint(project_root, artifact_dir, workflow_t
             step["stage_result"] = "REVIEW_PASS"
         if skill == "prizmkit-test":
             step["stage_result"] = "TEST_PASS"
-            step["authoritative_evidence_paths"] = relative_evidence
+            relative_artifact = artifact_dir.relative_to(project_root)
+            step["required_artifacts"] = [
+                str(relative_artifact / "test-report.md"),
+                str(relative_artifact / "test-result.json"),
+            ]
         steps.append(step)
         previous = step["id"]
     artifact_dir.mkdir(parents=True, exist_ok=True)
+    (artifact_dir / "test-report.md").write_text(
+        "# Test Report\n\n## Final Result\n\n- Result: TEST_PASS\n",
+        encoding="utf-8",
+    )
+    (artifact_dir / "test-result.json").write_text(
+        json.dumps({
+            "schema_version": 1,
+            "result": "TEST_PASS",
+            "report": "test-report.md",
+            "main_review_rounds": 1,
+            "independent_review": {
+                "status": "downgraded",
+                "responses": 0,
+                "downgrade_reason": "read-only continuation capability unavailable",
+                "final_state_rechecked": False,
+            },
+            "repair_rounds": 0,
+            "production_changed": False,
+            "review_required": False,
+            "review_scope": None,
+            "unresolved_items": [],
+        }),
+        encoding="utf-8",
+    )
     checkpoint = artifact_dir / "workflow-checkpoint.json"
     checkpoint.write_text(
         json.dumps({
@@ -169,7 +172,6 @@ def _write_semantic_completion_checkpoint(project_root, artifact_dir, workflow_t
             "steps": steps,
             state_key: {
                 "repair_round": 0,
-                "authoritative_evidence_paths": relative_evidence,
                 "terminal_status": "WORKFLOW_COMPLETED",
             },
         }),
@@ -3424,7 +3426,150 @@ def _terminal_success_result(tmp_path):
     )
 
 
-def test_invalid_existing_checkpoint_blocks_terminal_success_with_completion_summary(tmp_path):
+def test_session_status_parser_preserves_current_runtime_outcomes():
+    import importlib.util
+
+    script = PIPELINE_ROOT / "scripts" / "check-session-status.py"
+    spec = importlib.util.spec_from_file_location("check_session_status_contract", script)
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+
+    for status in (
+        "infra_error",
+        "stalled_context_continuation",
+        "workflow_blocked",
+        "workflow_skipped",
+    ):
+        assert module.determine_status({"status": status}) == status
+
+
+def test_terminal_success_without_semantic_completion_is_skipped(tmp_path):
+    from prizmkit_runtime.runner_classification import classify_session
+    from prizmkit_runtime.runner_models import PromptGenerationResult
+
+    base_head = _init_git_repo_for_invalid_checkpoint_test(tmp_path)
+    artifact_dir = tmp_path / ".prizmkit" / "specs" / "001-incomplete"
+    checkpoint = _write_semantic_completion_checkpoint(tmp_path, artifact_dir)
+    payload = json.loads(checkpoint.read_text(encoding="utf-8"))
+    payload["steps"][2]["required_artifacts"] = [
+        ".prizmkit/specs/001-incomplete/test-report.md",
+        ".prizmkit/specs/001-incomplete/missing-test-result.json",
+    ]
+    checkpoint.write_text(json.dumps(payload), encoding="utf-8")
+    prompt = PromptGenerationResult(
+        output_path=tmp_path / "prompt.md",
+        checkpoint_path=checkpoint,
+        artifact_path=artifact_dir,
+    )
+
+    classification = classify_session(
+        _terminal_success_result(tmp_path),
+        project_root=tmp_path,
+        base_head=base_head,
+        prompt=prompt,
+    )
+
+    assert classification.session_status == "workflow_skipped"
+    assert classification.reason == "no_semantic_completion:incomplete_semantic_evidence"
+    assert classification.fatal_error_code == ""
+
+
+def test_terminal_success_text_does_not_hide_provider_failure(tmp_path):
+    from prizmkit_runtime.runner_classification import classify_session
+    from prizmkit_runtime.runner_models import PromptGenerationResult
+    from prizmkit_runtime.sessions import AISessionCommand, AISessionResult
+    from prizmkit_runtime.status import ProgressSummary
+
+    base_head = _init_git_repo_for_invalid_checkpoint_test(tmp_path)
+    artifact_dir = tmp_path / ".prizmkit" / "specs" / "001-provider-error"
+    checkpoint = _write_semantic_completion_checkpoint(tmp_path, artifact_dir)
+    payload = json.loads(checkpoint.read_text(encoding="utf-8"))
+    payload["steps"][2]["required_artifacts"] = [
+        ".prizmkit/specs/001-provider-error/test-report.md",
+        ".prizmkit/specs/001-provider-error/missing-test-result.json",
+    ]
+    checkpoint.write_text(json.dumps(payload), encoding="utf-8")
+    log = tmp_path / "provider-error.log"
+    log.write_text("gateway failure after output", encoding="utf-8")
+    raw = AISessionResult(
+        command=AISessionCommand(("fake",), None, "claude", "stdin"),
+        exit_code=0,
+        timed_out=False,
+        termination_reason=None,
+        stale_marker=None,
+        fatal_error_code="provider_error",
+        progress_summary=ProgressSummary(result_subtype="success"),
+        session_log=log,
+        backup_log=tmp_path / "backup.log",
+        log_recovery=None,
+        started_epoch=0,
+        ended_epoch=1,
+    )
+
+    classification = classify_session(
+        raw,
+        project_root=tmp_path,
+        base_head=base_head,
+        prompt=PromptGenerationResult(
+            output_path=tmp_path / "prompt.md",
+            checkpoint_path=checkpoint,
+            artifact_path=artifact_dir,
+        ),
+    )
+
+    assert classification.session_status == "infra_error"
+    assert classification.reason == "ai_runtime_or_provider_error"
+
+
+@pytest.mark.parametrize(
+    ("kind", "item_id"),
+    [("feature", "F-001"), ("bugfix", "B-001"), ("refactor", "R-001")],
+)
+def test_family_status_skips_semantic_completion_failure_without_retry_spend(
+    tmp_path, kind, item_id
+):
+    from prizmkit_runtime.runner_models import family_for, parse_invocation
+    from prizmkit_runtime.runner_status import start_item, update_item
+
+    paths = _make_paths(tmp_path / kind)
+    item = {
+        "id": item_id,
+        "title": "Skip invalid completion evidence",
+        "status": "pending",
+        "priority": "high",
+        "dependencies": [],
+    }
+    if kind == "bugfix":
+        item.update(severity="high", description="invalid evidence", reproduction_steps=["run"])
+    elif kind == "refactor":
+        item.update(description="invalid evidence", targets=["src"], goals=["preserve"])
+    _write_family_plan(paths, kind, [item])
+    family = family_for(kind, paths)
+    invocation = parse_invocation(family, "run", ())
+    invocation = invocation.__class__(**{**invocation.__dict__, "list_path": family.plan_path})
+    assert start_item(family, invocation, item_id, paths.project_root).ok is True
+
+    result = update_item(
+        family,
+        invocation,
+        item_id,
+        "workflow_skipped",
+        "session-1",
+        paths.project_root,
+        checkpoint_state={},
+    )
+    status = json.loads((family.state_dir / item_id / "status.json").read_text(encoding="utf-8"))
+
+    assert result.ok is True
+    assert result.data["new_status"] == "skipped"
+    assert result.data["restart_policy"] == "terminal_skipped"
+    assert status["retry_count"] == 0
+    assert status["infra_error_count"] == 0
+    assert status["needs_attention"] is True
+    assert status["skipped_reason"] == "workflow_completion_evidence_invalid"
+
+
+def test_invalid_existing_checkpoint_skips_terminal_success_with_completion_summary(tmp_path):
     from prizmkit_runtime.runner_classification import classify_session
     from prizmkit_runtime.runner_models import PromptGenerationResult
 
@@ -3438,11 +3583,11 @@ def test_invalid_existing_checkpoint_blocks_terminal_success_with_completion_sum
 
     classification = classify_session(_terminal_success_result(tmp_path), project_root=tmp_path, base_head=base_head, prompt=prompt)
 
-    assert classification.session_status == "crashed"
+    assert classification.session_status == "workflow_skipped"
     assert classification.reason == "invalid_workflow_checkpoint:invalid_status"
 
 
-def test_invalid_existing_checkpoint_blocks_head_change_success(tmp_path):
+def test_invalid_existing_checkpoint_blocks_runtime_failure_with_head_change(tmp_path):
     from prizmkit_runtime.runner_classification import classify_session
     from prizmkit_runtime.runner_models import PromptGenerationResult
     from prizmkit_runtime.sessions import AISessionCommand, AISessionResult
@@ -3479,7 +3624,7 @@ def test_invalid_existing_checkpoint_blocks_head_change_success(tmp_path):
     assert classification.reason == "invalid_workflow_checkpoint:malformed_step_id"
 
 
-def test_invalid_existing_checkpoint_blocks_workspace_auto_commit_success(tmp_path):
+def test_invalid_existing_checkpoint_blocks_ambiguous_clean_exit(tmp_path):
     from prizmkit_runtime.runner_classification import classify_session
     from prizmkit_runtime.runner_models import PromptGenerationResult
     from prizmkit_runtime.sessions import AISessionCommand, AISessionResult
@@ -3877,11 +4022,6 @@ def test_family_status_persists_identical_semantic_blocked_snapshot_without_retr
         "stage_result": "TEST_BLOCKED",
         "repair_scope": "production",
         "repair_round": 2,
-        "authoritative_evidence_paths": {
-            "manifest": ".prizmkit/test/evidence/id/manifest.json",
-            "verdict": ".prizmkit/test/evidence/id/verdict.json",
-            "validation": ".prizmkit/test/evidence/id/validation.json",
-        },
         "blocked_reason": "test_blocked",
         "terminal_status": "WORKFLOW_BLOCKED",
     }

package/bundled/skills/_metadata.json

@@ -1,5 +1,5 @@
 {
-  "version": "1.1.140",
+  "version": "1.1.142",
   "skills": {
     "prizmkit": {
       "description": "Framework introduction and navigation for the formal single-requirement lifecycle, project initialization, Prizm docs, and independent deployment.",
@@ -9,7 +9,7 @@
       "hasScripts": false
     },
     "prizmkit-workflow": {
-      "description": "Composite L1 workflow for one explicitly requested formal requirement: plan → implement → code-review → test → retrospective → committer, with stable artifact identity and deterministic repair routing.",
+      "description": "Composite workflow for one explicitly requested formal requirement: plan → implement → code-review → test → retrospective → committer, with stable artifact identity and deterministic repair routing.",
       "tier": "foundation",
       "category": "prizmkit-skill",
       "hasAssets": false,
@@ -44,7 +44,7 @@
       "hasScripts": true
     },
     "prizmkit-committer": {
-      "description": "Final formal-requirement commit gate: verifies all prior stages, confirms interactive commits or performs authorized L4-headless local commits, and never pushes automatically.",
+      "description": "Final formal-requirement commit gate: verifies all prior stages, confirms interactive commits or performs authorized headless pipeline commits, and never pushes automatically.",
       "tier": "1",
       "category": "prizmkit-skill",
       "hasAssets": false,
@@ -72,11 +72,11 @@
       "hasScripts": false
     },
     "prizmkit-test": {
-      "description": "Mandatory post-review auditable test evidence with TEST_PASS/TEST_FAIL/TEST_BLOCKED verdicts, repair-scope routing, current-workspace execution evidence, replay, and deterministic validation.",
+      "description": "AI-led project-native testing for the affected business module and Regression Ring, with bounded failure repair, mandatory Main-Agent review, optional strict independent Test Reviewer, and TEST_PASS/TEST_NEEDS_FIXES/TEST_BLOCKED results.",
       "tier": "1",
       "category": "prizmkit-skill",
-      "hasAssets": true,
-      "hasScripts": true
+      "hasAssets": false,
+      "hasScripts": false
     },
     "feature-workflow": {
       "description": "Thin feature scenario router. Clarifies whether the request needs direct single-requirement guidance, app planning, feature planning, or command construction from an existing feature list.",

package/bundled/skills/prizmkit/SKILL.md

@@ -69,12 +69,12 @@ Typos, pure formatting, small documentation edits, and other explicitly low-risk
 
 | Stage | Responsibility | Success handoff |
 |---|---|---|
-| `prizmkit-plan` | Clarify the requirement and create/review `spec.md` and `plan.md`. | `PLAN_READY` → `prizmkit-implement` |
-| `prizmkit-implement` | Execute the plan tasks and record completion. | `IMPLEMENTED` → `prizmkit-code-review` |
-| `prizmkit-code-review` | Main Agent reviews, repairs, verifies, and loops until convergence. | `REVIEW_PASS` → `prizmkit-test` |
-| `prizmkit-test` | Validate the final reviewed workspace with project-native evidence. | `TEST_PASS` → `prizmkit-retrospective` |
-| `prizmkit-retrospective` | Synchronize durable project documentation or record no documentation change. | `RETRO_COMPLETE` → `prizmkit-committer` |
-| `prizmkit-committer` | Verify gates, request commit confirmation, and create the local commit. | `COMMITTED` |
+| `prizmkit-plan` | Clarify the requirement and create/review `spec.md` and `plan.md`. | `status=completed`, `stage_result=PLAN_READY` → `prizmkit-implement` |
+| `prizmkit-implement` | Execute the plan tasks and record completion. | `status=completed`, `stage_result=IMPLEMENTED` → `prizmkit-code-review` |
+| `prizmkit-code-review` | Main Agent reviews, repairs, verifies, and loops until convergence. | `status=completed`, `stage_result=REVIEW_PASS` → `prizmkit-test` |
+| `prizmkit-test` | Validate the final reviewed workspace with project-native tests and a consistent terminal report/result pair. | `status=completed`, `stage_result=TEST_PASS` → `prizmkit-retrospective` |
+| `prizmkit-retrospective` | Synchronize durable project documentation or record no documentation change. | `status=completed`, `stage_result=RETRO_COMPLETE` → `prizmkit-committer` |
+| `prizmkit-committer` | Verify gates, request commit confirmation, and create the local commit. | `status=completed`, `stage_result=COMMITTED` |
 
 ### `prizmkit-workflow`
 
@@ -108,19 +108,14 @@ REVIEW_NEEDS_FIXES
   → prizmkit-implement
   → prizmkit-code-review
 
-TEST_FAIL affecting only tests, fixtures, or test-runner configuration
-  → prizmkit-implement
-  → prizmkit-test
-
-TEST_FAIL affecting production code, runtime configuration, schema,
-dependencies, or public interfaces
-  → prizmkit-implement
-  → prizmkit-code-review
-  → prizmkit-test
+TEST_NEEDS_FIXES
+  → preserve test-report.md and test-result.json
+  → stop with the known correction or delta-review requirement
+  → caller owns any later review/retest route
 
 TEST_BLOCKED
-  → pause without speculative code changes
-  → resume from prizmkit-test when the environment is available
+  → preserve the blocker without speculative code changes
+  → caller owns recovery after truth, input, safety, environment, or reliability is restored
 ```
 
 Automatic outer repair is limited to three rounds. The workflow stops with a resumable blocked result when the limit is reached or a gate cannot be safely completed.
@@ -141,7 +136,7 @@ An active requirement may use:
 
 This runtime state records the current stage, terminal status, repair scope, repair round, and resume entry. The skill set does not prescribe whether the target project commits, ignores, or shares this file.
 
-Read `${SKILL_DIR}/references/workflow-state-protocol.md` for the state protocol. If the state file is missing or stale, use `spec.md`, `plan.md`, review evidence, and test evidence to reconstruct the safest recoverable stage and report that reconstruction.
+Read `${SKILL_DIR}/references/workflow-state-protocol.md` for the state protocol. If the state file is missing or stale, use `spec.md`, `plan.md`, the review report, `test-report.md`, and `test-result.json` to reconstruct the safest recoverable stage and report that reconstruction.
 
 ## Quick Start