prizmkit 1.1.141 → 1.1.143

package/bundled/dev-pipeline/templates/bootstrap-tier3.md

@@ -237,7 +237,7 @@ For each Verification Gate from the Task Contract, write one evidence line to Im
 
 Review the completed implementation before the full Feature test gate. Run `/prizmkit-code-review artifact_dir=.prizmkit/specs/{{FEATURE_SLUG}}/` and follow the skill's current review contract for review execution, repairs, verification, evidence, and handoff. Do not invoke another review skill or add a second review path outside that contract.
 
-- `REVIEW_PASS` → record `stage=code-review`, `status=REVIEW_PASS`, `stage_result=PASS`, `next_stage=prizmkit-test`, and `resume_from=prizmkit-test`; then run the test gate.
+- `REVIEW_PASS` → record `stage=code-review`, `status=completed`, `stage_result=REVIEW_PASS`, `next_stage=prizmkit-test`, and `resume_from=prizmkit-test`; then run the test gate.
 - `REVIEW_NEEDS_FIXES` → increment the shared outer `repair_round` and route `prizmkit-implement` → `prizmkit-code-review` → `prizmkit-test`.
 - Stop as `WORKFLOW_BLOCKED` when the outer repair round is already three. The Main-Agent review loop has its separate limit of up to ten completed rounds.
 

package/bundled/dev-pipeline/templates/sections/bugfix-phase-review.md

@@ -16,4 +16,4 @@ Missing required evidence, an unsafe repair, a failed verification, or failure t
 - `REVIEW_NEEDS_FIXES` → route through bounded implementation/review/test repair; do not commit.
 - `REVIEW_INCOMPLETE` or `REVIEW_INVALID` → inspect progress once, write `failure-log.md`, and stop. Do not accept old or incidental verdict text and do not synthesize a passing result outside the skill.
 
-**Checkpoint update**: Set step `prizmkit-code-review` to `completed` with `stage_result=REVIEW_PASS` only for an accepted valid Final Result.
+**Checkpoint update**: Set step `prizmkit-code-review` to `completed` with `stage_result=REVIEW_PASS` only for an accepted valid Final Result. Set it to `failed` with `stage_result=REVIEW_NEEDS_FIXES` when the review returns known unresolved findings.

package/bundled/dev-pipeline/templates/sections/checkpoint-system.md

@@ -19,8 +19,12 @@ Use this helper to update checkpoint status. Do not hand-edit `workflow-checkpoi
 
 - Use the step `skill` value such as `prizmkit-implement`, or the step id such as `S04`.
 - Omit `--stage-result` for stages without a domain result.
+- Complete `prizmkit-plan` with `--status completed --stage-result PLAN_READY`.
+- Complete `prizmkit-implement` with `--status completed --stage-result IMPLEMENTED`.
 - Complete `prizmkit-code-review` only with `--status completed --stage-result REVIEW_PASS`.
 - Complete `prizmkit-test` only after validating the sibling report/result pair, with `--status completed --stage-result TEST_PASS`.
+- Complete `prizmkit-retrospective` with `--status completed --stage-result RETRO_COMPLETE`.
+- Record commit authorization wait with `--status in_progress --stage-result COMMIT_PENDING`; after the local commit, use `--status completed --stage-result COMMITTED`.
 - For `TEST_NEEDS_FIXES` or `TEST_BLOCKED`, use `--status failed --stage-result <result>` and preserve its report/result and blocker; never mark the test step completed.
 - Omit `--note` when there is no useful failure or handoff note.
 
@@ -28,7 +32,7 @@ Use this helper to update checkpoint status. Do not hand-edit `workflow-checkpoi
 
 1. Before each skill, verify its predecessor is `completed` or `skipped`.
 2. Starting a skill marks its step `in_progress`; a valid success marks it `completed`; a non-pass records its truthful `stage_result` and leaves later stages incomplete.
-3. Pipeline metadata owns only `current_stage`, `next_stage`, `stage_result`, `repair_scope`, `repair_round`, `blocked_reason`, and `terminal_status`.
+3. Pipeline metadata owns only `current_stage`, `next_stage`, `stage_result`, `repair_scope`, `repair_round`, `blocked_reason`, and `terminal_status`; lifecycle `status` belongs to each checkpoint step.
 4. Skills own their reports and terminal results. The pipeline must not duplicate a skill's internal state machine or require evidence manifests, hashes, or attestations.
 5. Preserve all reports and session artifacts on recovery. Never replace skill artifacts with the pipeline checkpoint or merge their schemas.
 6. `TEST_NEEDS_FIXES` is a known testing-domain non-pass. `TEST_BLOCKED` records terminal `WORKFLOW_BLOCKED` with a deterministic reason. Neither is an AI CLI crash.

package/bundled/dev-pipeline/templates/sections/phase-review-full.md

@@ -4,8 +4,8 @@ Review the completed implementation before the full Feature test gate runs. A re
 
 Use `artifact_dir=.prizmkit/specs/{{FEATURE_SLUG}}/` so `/prizmkit-code-review` can read the spec, plan, context snapshot, implementation log, and current diff. Follow the skill's current review contract for review execution, repairs, verification, evidence, and handoff. Do not invoke another review skill or add a second review path outside that contract.
 
-- `REVIEW_PASS` → update checkpoint metadata with `stage=code-review`, `status=REVIEW_PASS`, `stage_result=REVIEW_PASS`, `next_stage=prizmkit-test`, and `resume_from=prizmkit-test`; then run the full test gate.
-- `REVIEW_NEEDS_FIXES` → preserve findings, increment the shared outer `repair_round`, route `prizmkit-implement` → `prizmkit-code-review` → `prizmkit-test`, and do not accept any test result from before the repaired review.
+- `REVIEW_PASS` → update checkpoint metadata with `stage=code-review`, `status=completed`, `stage_result=REVIEW_PASS`, `next_stage=prizmkit-test`, and `resume_from=prizmkit-test`; then run the full test gate.
+- `REVIEW_NEEDS_FIXES` → update the review step with `status=failed` and `stage_result=REVIEW_NEEDS_FIXES`, preserve findings, increment the shared outer `repair_round`, route `prizmkit-implement` → `prizmkit-code-review` → `prizmkit-test`, and do not accept any test result from before the repaired review.
 - If the outer repair round is already three, record `WORKFLOW_BLOCKED`; do not start another repair route.
 
 Missing required evidence, unsafe repair, failed verification, or failure to converge under the review contract produces `REVIEW_NEEDS_FIXES`.

package/bundled/dev-pipeline/templates/sections/refactor-phase-review.md

@@ -16,4 +16,4 @@ Missing required evidence, an unsafe repair, a failed verification, or failure t
 
 Verify directly that the review checks behavior preservation, public API compatibility, imports, tests, and rollback assumptions.
 
-**Checkpoint update**: Set step `prizmkit-code-review` to `completed` with `stage_result=REVIEW_PASS` only for an accepted valid Final Result. Do not complete the checkpoint for incomplete or invalid reports.
+**Checkpoint update**: Set step `prizmkit-code-review` to `completed` with `stage_result=REVIEW_PASS` only for an accepted valid Final Result. Set it to `failed` with `stage_result=REVIEW_NEEDS_FIXES` when the review returns known unresolved findings. Do not complete the checkpoint for incomplete or invalid reports.

package/bundled/dev-pipeline/templates/session-status-schema.json

@@ -15,7 +15,7 @@
     },
     "status": {
       "type": "string",
-      "enum": ["success", "partial", "failed", "commit_missing", "docs_missing", "merge_conflict", "context_overflow"]
+      "enum": ["success", "partial", "failed", "crashed", "timed_out", "infra_error", "commit_missing", "docs_missing", "merge_conflict", "context_overflow", "stalled_context_continuation", "workflow_blocked", "workflow_skipped"]
     },
     "completed_phases": {
       "type": "array",

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

@@ -5,6 +5,9 @@ from __future__ import annotations
 import json
 
 from prizmkit_runtime.checkpoint_state import (
+    VALID_CHECKPOINT_STATUSES,
+    VALID_STAGE_RESULTS,
+    expected_status_for_stage_result,
     load_checkpoint_state,
     read_checkpoint_cursor,
     validate_checkpoint_data,
@@ -143,6 +146,169 @@ def test_structural_validation_error_codes(tmp_path):
         assert read_checkpoint_cursor(path) is None
 
 
+def test_domain_results_map_to_lifecycle_statuses():
+    expected = {
+        "PLAN_READY": "completed",
+        "PLAN_BLOCKED": "failed",
+        "IMPLEMENTED": "completed",
+        "IMPLEMENT_REPAIR": "failed",
+        "IMPLEMENT_BLOCKED": "failed",
+        "REVIEW_PASS": "completed",
+        "REVIEW_NEEDS_FIXES": "failed",
+        "TEST_PASS": "completed",
+        "TEST_NEEDS_FIXES": "failed",
+        "TEST_BLOCKED": "failed",
+        "RETRO_COMPLETE": "completed",
+        "RETRO_BLOCKED": "failed",
+        "COMMIT_PENDING": "in_progress",
+        "COMMITTED": "completed",
+        "COMMIT_BLOCKED": "failed",
+    }
+
+    assert VALID_CHECKPOINT_STATUSES == {
+        "pending", "in_progress", "completed", "failed", "skipped",
+    }
+    for result, status in expected.items():
+        assert expected_status_for_stage_result(result) == status
+
+
+def test_structural_validation_rejects_domain_result_as_status(tmp_path):
+    for domain_status in [
+        "PLAN_READY", "IMPLEMENTED", "REVIEW_PASS", "REVIEW_NEEDS_FIXES",
+        "TEST_PASS", "TEST_NEEDS_FIXES", "TEST_BLOCKED", "RETRO_COMPLETE",
+        "COMMIT_PENDING", "COMMITTED",
+    ]:
+        state = validate_checkpoint_data(
+            _checkpoint([_step("S01", status=domain_status)]),
+            path=tmp_path / f"{domain_status}.json",
+        )
+        assert state.valid is False
+        assert state.error_code == "invalid_status"
+
+
+def test_structural_validation_rejects_status_result_mismatch(tmp_path):
+    step = _step("S01", "prizmkit-code-review", "Review", "failed")
+    step["stage_result"] = "REVIEW_PASS"
+
+    state = validate_checkpoint_data(
+        _checkpoint([step]),
+        path=tmp_path / "mismatch.json",
+    )
+
+    assert state.valid is False
+    assert state.error_code == "stage_result_status_mismatch"
+
+
+def test_structural_validation_rejects_stage_result_owner_mismatch(tmp_path):
+    for skill in ("prizmkit-test", "custom-stage"):
+        step = _step("S01", skill, "Step", "completed")
+        step["stage_result"] = "REVIEW_PASS"
+
+        state = validate_checkpoint_data(
+            _checkpoint([step]),
+            path=tmp_path / f"{skill}-owner-mismatch.json",
+        )
+
+        assert state.valid is False
+        assert state.error_code == "stage_result_owner_mismatch"
+
+
+def test_semantic_metadata_rejects_legacy_or_mismatched_status(tmp_path):
+    invalid = [
+        ({"status": "TEST_PASS", "stage_result": "TEST_PASS"}, "invalid_semantic_status"),
+        ({"status": "failed", "stage_result": "TEST_PASS"}, "semantic_stage_result_status_mismatch"),
+        ({"status": "completed", "stage_result": "UNKNOWN_RESULT"}, "invalid_semantic_stage_result"),
+        ({"stage": "test", "status": "completed", "stage_result": "REVIEW_PASS"}, "semantic_stage_result_owner_mismatch"),
+    ]
+
+    for index, (semantic, error_code) in enumerate(invalid):
+        state = validate_checkpoint_data(
+            _checkpoint([_step("S01")]) | {"feature_state": semantic},
+            path=tmp_path / f"semantic-status-{index}.json",
+        )
+
+        assert state.valid is True
+        assert state.semantic.error_code == error_code
+
+
+def test_update_checkpoint_reenters_failed_stage_without_stale_result(tmp_path):
+    payload = _family_checkpoint(
+        tmp_path,
+        semantic={
+            "stage": "code-review",
+            "status": "failed",
+            "stage_result": "REVIEW_NEEDS_FIXES",
+            "repair_scope": "production",
+            "repair_round": 1,
+        },
+        terminal=False,
+    )
+    payload["steps"][1]["status"] = "failed"
+    payload["steps"][1]["stage_result"] = "REVIEW_NEEDS_FIXES"
+    for step in payload["steps"][2:]:
+        step["status"] = "pending"
+        step.pop("stage_result", None)
+    checkpoint = _write(tmp_path / "workflow-checkpoint.json", payload)
+
+    result = update_checkpoint(
+        str(checkpoint),
+        "prizmkit-code-review",
+        "in_progress",
+    )
+    persisted = json.loads(checkpoint.read_text(encoding="utf-8"))
+
+    assert result["ok"] is True
+    assert result["semantic"]["current_stage"] == "prizmkit-code-review"
+    assert result["semantic"]["next_stage"] == "prizmkit-code-review"
+    assert persisted["steps"][1]["status"] == "in_progress"
+    assert "stage_result" not in persisted["steps"][1]
+    assert persisted["feature_state"]["stage"] == "code-review"
+    assert persisted["feature_state"]["status"] == "in_progress"
+    assert "stage_result" not in persisted["feature_state"]
+
+
+def test_checkpoint_updater_supports_every_owned_stage_result(tmp_path):
+    cases = [
+        ("prizmkit-plan", "PLAN_READY", "completed"),
+        ("prizmkit-implement", "IMPLEMENTED", "completed"),
+        ("prizmkit-code-review", "REVIEW_PASS", "completed"),
+        ("prizmkit-test", "TEST_BLOCKED", "failed"),
+        ("prizmkit-retrospective", "RETRO_COMPLETE", "completed"),
+        ("prizmkit-committer", "COMMIT_PENDING", "in_progress"),
+        ("prizmkit-committer", "COMMITTED", "completed"),
+    ]
+
+    assert {stage_result for _skill, stage_result, _status in cases} <= VALID_STAGE_RESULTS
+    for index, (skill, stage_result, status) in enumerate(cases):
+        step = _step("S01", skill, skill, "pending")
+        checkpoint = _write(
+            tmp_path / f"owned-result-{index}.json",
+            _checkpoint([step]),
+        )
+
+        result = update_checkpoint(
+            str(checkpoint),
+            skill,
+            status,
+            semantic_update={"stage_result": stage_result},
+        )
+
+        assert result["ok"] is True
+        persisted = json.loads(checkpoint.read_text(encoding="utf-8"))
+        assert persisted["steps"][0]["status"] == status
+        assert persisted["steps"][0]["stage_result"] == stage_result
+        assert persisted["feature_state"]["stage"] == {
+            "prizmkit-plan": "plan",
+            "prizmkit-implement": "implement",
+            "prizmkit-code-review": "code-review",
+            "prizmkit-test": "test",
+            "prizmkit-retrospective": "retrospective",
+            "prizmkit-committer": "committer",
+        }[skill]
+        assert persisted["feature_state"]["status"] == status
+        assert persisted["feature_state"]["stage_result"] == stage_result
+
+
 def test_duplicate_skills_require_exact_step_id(tmp_path):
     path = _write(tmp_path / "workflow-checkpoint.json", _checkpoint([
         _step("S01", "prizmkit-committer", "First", "completed"),
@@ -305,6 +471,7 @@ def test_update_checkpoint_rejects_test_before_review_pass(tmp_path):
     payload["steps"][1]["status"] = "skipped"
     payload["steps"][1].pop("stage_result", None)
     payload["steps"][2]["status"] = "pending"
+    payload["steps"][2].pop("stage_result", None)
     for step in payload["steps"][3:]:
         step["status"] = "pending"
         step.pop("stage_result", None)
@@ -391,6 +558,42 @@ def test_completion_rejects_inconsistent_independent_review_metadata(tmp_path):
     assert state.semantic.artifacts_valid is False
 
 
+def test_completion_accepts_legacy_terminal_result_report(tmp_path):
+    payload = _family_checkpoint(tmp_path)
+    checkpoint = _write(
+        tmp_path / ".prizmkit/specs/feature/workflow-checkpoint.json",
+        payload,
+    )
+    (checkpoint.parent / "test-report.md").write_text(
+        "# Test Report\n\n## Terminal Result\n\n- Result: TEST_PASS\n\n"
+        "## Final Decision\n\nTEST_PASS\n",
+        encoding="utf-8",
+    )
+
+    state = load_checkpoint_state(checkpoint, project_root=tmp_path)
+
+    assert state.semantic_complete is True
+    assert state.semantic.artifacts_valid is True
+
+
+def test_completion_rejects_conflicting_supported_report_results(tmp_path):
+    payload = _family_checkpoint(tmp_path)
+    checkpoint = _write(
+        tmp_path / ".prizmkit/specs/feature/workflow-checkpoint.json",
+        payload,
+    )
+    (checkpoint.parent / "test-report.md").write_text(
+        "# Test Report\n\n## Terminal Result\n\n- Result: TEST_PASS\n\n"
+        "## Final Decision\n\nTEST_BLOCKED\n",
+        encoding="utf-8",
+    )
+
+    state = load_checkpoint_state(checkpoint, project_root=tmp_path)
+
+    assert state.semantic_complete is False
+    assert state.semantic.artifacts_valid is False
+
+
 def test_completion_rejects_mismatched_test_report_result(tmp_path):
     payload = _family_checkpoint(tmp_path)
     checkpoint = _write(
@@ -447,6 +650,34 @@ def test_completion_rejects_test_artifacts_outside_project(tmp_path):
     assert state.semantic.artifacts_valid is False
 
 
+def test_review_needs_fixes_preserves_implementation_repair_route(tmp_path):
+    payload = _family_checkpoint(
+        tmp_path,
+        semantic={"repair_scope": "production", "repair_round": 0},
+        terminal=False,
+    )
+    payload["steps"][1]["status"] = "in_progress"
+    payload["steps"][1].pop("stage_result")
+    for step in payload["steps"][2:]:
+        step["status"] = "pending"
+        step.pop("stage_result", None)
+    checkpoint = _write(tmp_path / "workflow-checkpoint.json", payload)
+
+    result = update_checkpoint(
+        str(checkpoint),
+        "prizmkit-code-review",
+        "failed",
+        semantic_update={"stage_result": "REVIEW_NEEDS_FIXES", "repair_scope": "production"},
+    )
+    persisted = json.loads(checkpoint.read_text(encoding="utf-8"))
+
+    assert result["ok"] is True
+    assert result["semantic"]["next_stage"] == "prizmkit-implement"
+    assert persisted["feature_state"]["repair_round"] == 1
+    assert persisted["feature_state"]["current_stage"] == "prizmkit-code-review"
+    assert persisted["feature_state"]["next_stage"] == "prizmkit-implement"
+
+
 def test_review_outer_repair_cap_remains_runtime_owned(tmp_path):
     payload = _family_checkpoint(
         tmp_path, semantic={"repair_scope": "production", "repair_round": 3},

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

@@ -31,6 +31,7 @@ from generate_bootstrap_prompt import (
     main as generate_bootstrap_main,
 )
 from continuation import append_continuation_handoff, checkpoint_cursor
+from prompt_framework import pipeline_checkpoint_metadata
 from utils import resolve_prompt_platform
 
 REMOVED_MAX_LOG_ENV = "MAX_" + "LOG_SIZE"
@@ -472,8 +473,13 @@ class TestScopedFeatureTestGate:
         assert "{{CHECKPOINT_PYTHON_CMD}}" in rendered
         assert "--checkpoint-path {{CHECKPOINT_PATH}}" in rendered
         assert "--stage-result <optional-domain-result>" in rendered
+        assert "--status completed --stage-result PLAN_READY" in rendered
+        assert "--status completed --stage-result IMPLEMENTED" in rendered
         assert "--status completed --stage-result REVIEW_PASS" in rendered
         assert "--status completed --stage-result TEST_PASS" in rendered
+        assert "--status completed --stage-result RETRO_COMPLETE" in rendered
+        assert "--status in_progress --stage-result COMMIT_PENDING" in rendered
+        assert "--status completed --stage-result COMMITTED" in rendered
         assert "Do not hand-edit `workflow-checkpoint.json` directly." in rendered
         assert "authoritative_evidence_paths" not in rendered
 
@@ -562,14 +568,33 @@ class TestScopedFeatureTestGate:
         assert blocked["blocked_reason"] == "test_blocked"
         assert blocked["terminal_status"] == "WORKFLOW_BLOCKED"
 
+    def test_pipeline_checkpoint_metadata_rejects_legacy_or_mismatched_status(self):
+        invalid = [
+            {"status": "TEST_PASS", "stage_result": "TEST_PASS"},
+            {"status": "failed", "stage_result": "TEST_PASS"},
+            {"status": "completed", "stage_result": "UNKNOWN_RESULT"},
+        ]
+
+        for values in invalid:
+            try:
+                pipeline_checkpoint_metadata(
+                    ".prizmkit/specs/contract-test",
+                    stage="test",
+                    **values,
+                )
+            except ValueError:
+                continue
+            raise AssertionError("invalid lifecycle metadata was accepted: {}".format(values))
+
     def test_feature_checkpoint_metadata_has_no_evidence_paths(self):
         metadata = feature_checkpoint_metadata(
             ".prizmkit/specs/123-scope-test",
-            stage="test", status="TEST_NEEDS_FIXES",
+            stage="test", status="failed",
             stage_result="TEST_NEEDS_FIXES", repair_round=0,
             next_stage=None, resume_from=None,
         )
 
+        assert metadata["status"] == "failed"
         assert metadata["stage_result"] == "TEST_NEEDS_FIXES"
         assert "authoritative_evidence_paths" not in metadata
         assert metadata["artifact_dir"] == ".prizmkit/specs/123-scope-test"

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"
 
@@ -527,7 +528,7 @@ class TestBugfixCheckpointGeneration:
         metadata = bugfix_checkpoint_metadata(
             ".prizmkit/bugfix/B-001",
             stage="test",
-            status="TEST_NEEDS_FIXES",
+            status="failed",
             stage_result="TEST_NEEDS_FIXES",
             repair_round=0,
             next_stage=None,
@@ -535,6 +536,7 @@ class TestBugfixCheckpointGeneration:
         )
 
         assert metadata["artifact_dir"] == ".prizmkit/bugfix/B-001"
+        assert metadata["status"] == "failed"
         assert metadata["stage_result"] == "TEST_NEEDS_FIXES"
         assert "authoritative_evidence_paths" not in metadata
 

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

@@ -59,7 +59,7 @@ class TestRefactorCheckpointMerge:
         existing["refactor_state"] = refactor_checkpoint_metadata(
             ".prizmkit/refactor/R-008",
             stage="test",
-            status="TEST_NEEDS_FIXES",
+            status="failed",
             stage_result="TEST_NEEDS_FIXES",
             repair_scope="public-interface",
             repair_round=2,
@@ -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
 
@@ -376,13 +377,15 @@ class TestRefactorCheckpointGeneration:
         metadata = refactor_checkpoint_metadata(
             ".prizmkit/refactor/R-001",
             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["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"

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

@@ -3426,7 +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
 
@@ -3440,11 +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
@@ -3481,7 +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

package/bundled/skills/_metadata.json

@@ -1,5 +1,5 @@
 {
-  "version": "1.1.141",
+  "version": "1.1.143",
   "skills": {
     "prizmkit": {
       "description": "Framework introduction and navigation for the formal single-requirement lifecycle, project initialization, Prizm docs, and independent deployment.",