clawed 2.3.7__tar.gz → 2.3.8__tar.gz

{clawed-2.3.7 → clawed-2.3.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: clawed
-Version: 2.3.7
+Version: 2.3.8
 Summary: Claw-ED — personal AI teaching agent. Learns your voice, works while you sleep.
 Project-URL: Homepage, https://github.com/SirhanMacx/Claw-ED
 Project-URL: Documentation, https://github.com/SirhanMacx/Claw-ED#readme
@@ -82,6 +82,18 @@ Built on the OpenClaw agent framework. Open source. MIT license.
 
 ---
 
+## What's new in v2.3.8
+
+**No more silent failures.** Every step of lesson generation now reports what happened — persona loading, material search, quality review, voice matching. If something fails, you'll know exactly what and why, with structured NLAH failure codes. Quality review runs automatically on every generated lesson and fails closed: if the review itself crashes, it reports failure instead of silently passing.
+
+**Stricter quality gates.** Lessons now require at least 6 guided notes, 3 exit ticket questions, and 2 primary sources with actual text. Topic drift is caught automatically. Voice match scoring compares generated lessons against your teaching persona.
+
+**Safer onboarding.** Teacher names and subjects are validated and truncated. Invalid grade levels get a re-prompt instead of being silently accepted. Demo mode can be forced with `CLAWED_DEMO=1` for presentations.
+
+**Async cleanup.** Background ingestion no longer risks crashing on Python 3.12+ from nested event loops. The fix is shared across all async call sites.
+
+---
+
 ## What's new in v2.3.7
 
 **Real images in every lesson.** Image specs are now required for every primary source and instruction section across all subjects. The LLM generates specific search queries ("Thomas Nast Boss Tweed political cartoon 1871") instead of leaving the field blank. Teacher images are found first using a three-stage progressive search (full query, individual keywords, subject fallback) with filename-weighted scoring across up to 150 candidates. External sources (Library of Congress, Wikimedia Commons, Unsplash) fill in the rest with subject-aware routing.

{clawed-2.3.7 → clawed-2.3.8}/README.md

@@ -13,6 +13,18 @@ Built on the OpenClaw agent framework. Open source. MIT license.
 
 ---
 
+## What's new in v2.3.8
+
+**No more silent failures.** Every step of lesson generation now reports what happened — persona loading, material search, quality review, voice matching. If something fails, you'll know exactly what and why, with structured NLAH failure codes. Quality review runs automatically on every generated lesson and fails closed: if the review itself crashes, it reports failure instead of silently passing.
+
+**Stricter quality gates.** Lessons now require at least 6 guided notes, 3 exit ticket questions, and 2 primary sources with actual text. Topic drift is caught automatically. Voice match scoring compares generated lessons against your teaching persona.
+
+**Safer onboarding.** Teacher names and subjects are validated and truncated. Invalid grade levels get a re-prompt instead of being silently accepted. Demo mode can be forced with `CLAWED_DEMO=1` for presentations.
+
+**Async cleanup.** Background ingestion no longer risks crashing on Python 3.12+ from nested event loops. The fix is shared across all async call sites.
+
+---
+
 ## What's new in v2.3.7
 
 **Real images in every lesson.** Image specs are now required for every primary source and instruction section across all subjects. The LLM generates specific search queries ("Thomas Nast Boss Tweed political cartoon 1871") instead of leaving the field blank. Teacher images are found first using a three-stage progressive search (full query, individual keywords, subject fallback) with filename-weighted scoring across up to 150 candidates. External sources (Library of Congress, Wikimedia Commons, Unsplash) fill in the rest with subject-aware routing.

{clawed-2.3.7 → clawed-2.3.8}/clawed/__init__.py

@@ -17,7 +17,7 @@ if hasattr(sys.stderr, "reconfigure"):
     except Exception:
         pass
 
-__version__ = "2.3.7"
+__version__ = "2.3.8"
 __author__ = "Jon Maccarello & Claw-ED contributors"
 __description__ = "Personal AI teaching agent. Learns your voice, works while you sleep."
 

{clawed-2.3.7 → clawed-2.3.8}/clawed/agent_core/core.py

@@ -339,7 +339,9 @@ class Gateway:
         # 2a. Load SOUL.md if available
         soul_context = ""
         try:
-            soul_path = Path.home() / ".eduagent" / "workspace" / "SOUL.md"
+            import os
+            data_dir = os.environ.get("EDUAGENT_DATA_DIR", str(Path.home() / ".eduagent"))
+            soul_path = Path(data_dir) / "workspace" / "SOUL.md"
             if soul_path.exists():
                 soul_context = soul_path.read_text(encoding="utf-8")[:2000]
         except Exception:

{clawed-2.3.7 → clawed-2.3.8}/clawed/agent_core/tools/generate_lesson_bundle.py

@@ -6,6 +6,7 @@ from pathlib import Path
 from typing import Any
 
 from clawed.agent_core.context import AgentContext, ToolResult
+from clawed.failure_codes import FailureCode
 
 logger = logging.getLogger(__name__)
 
@@ -89,14 +90,18 @@ class GenerateLessonBundleTool:
             f"this usually takes 2-4 minutes. I'll send everything when it's ready!"
         )
 
+        from clawed.generation_report import GenerationReport
+        report = GenerationReport()
+
         # ── Load config & persona from context ───────────────────────
         config = context.config
         persona = TeacherPersona()
         if context.persona:
             try:
                 persona = TeacherPersona(**context.persona)
-            except Exception:
-                pass
+            except Exception as e:
+                logger.warning("NLAH_FAILURE=%s: %s", FailureCode.PERSONA_PARSE_ERROR, e)
+                report.warnings.append(f"[{FailureCode.PERSONA_PARSE_ERROR}] Could not parse persona: {e}")
 
         # ── Load state standards if teacher profile has a state ───────
         state = ""
@@ -127,7 +132,8 @@ class GenerateLessonBundleTool:
                     len(assets), len(yt_links), topic,
                 )
         except Exception as e:
-            logger.debug("Asset search failed: %s", e)
+            logger.warning("NLAH_FAILURE=%s: %s", FailureCode.ASSET_SEARCH_FAILED, e)
+            report.warnings.append(f"[{FailureCode.ASSET_SEARCH_FAILED}] Asset search failed: {e}")
 
         # KB chunk-level search (text excerpts)
         try:
@@ -162,7 +168,8 @@ class GenerateLessonBundleTool:
                         )
                     logger.info("KB search found %d relevant chunks for '%s'", len(kb_parts), topic)
         except Exception as e:
-            logger.debug("KB search failed: %s", e)
+            logger.warning("NLAH_FAILURE=%s: %s", FailureCode.KB_SEARCH_FAILED, e)
+            report.warnings.append(f"[{FailureCode.KB_SEARCH_FAILED}] KB search failed: {e}")
 
         # ── Build a UnitPlan with standards ──────────────────────────
         description = f"Introduction to {topic}"
@@ -205,14 +212,12 @@ class GenerateLessonBundleTool:
                 teacher_materials=kb_prompt_section,
             )
         except Exception as e:
-            return ToolResult(text=f"Failed to generate lesson: {e}")
+            logger.error("NLAH_FAILURE=%s: %s", FailureCode.API_FAILURE, e)
+            return ToolResult(text=f"[{FailureCode.API_FAILURE}] Failed to generate lesson: {type(e).__name__}")
 
         # ── Validate ──────────────────────────────────────────────────
-        from clawed.generation_report import GenerationReport
         from clawed.validation import check_self_contained, validate_alignment, validate_master_content
 
-        report = GenerationReport()
-
         mc_errors = validate_master_content(master, topic)
         for err in mc_errors:
             report.warnings.append(err)
@@ -283,6 +288,45 @@ class GenerateLessonBundleTool:
             logger.error("Slides compile failed: %s", e)
             errors.append(f"Slideshow PPTX failed: {e}")
 
+        # ── Quality review (NLAH Stage 4 — non-blocking) ────────────
+        try:
+            from clawed.llm import LLMClient
+            from clawed.quality import score_voice_match
+
+            llm = LLMClient(config=config)
+            master_json = master.model_dump_json(indent=2)[:3000]
+            review = await llm.review_lesson_package(
+                lesson_json=master_json,
+                standards_present=bool(standards_list),
+                has_handout=len(generated_files) >= 2,
+                has_slideshow=len(generated_files) >= 3,
+            )
+            report.quality_review_passed = review.get("passed", False)
+            report.quality_review_issues = review.get("issues", [])
+            if not report.quality_review_passed:
+                for issue in report.quality_review_issues:
+                    report.warnings.append(f"[REVIEW] {issue}")
+                logger.info("Quality review: FAILED — %d issues", len(report.quality_review_issues))
+            else:
+                logger.info("Quality review: PASSED")
+
+            # Voice match scoring
+            persona_ctx = persona.to_prompt_context() if persona else ""
+            all_text = " ".join(s.content for s in master.direct_instruction)
+            voice_score = await score_voice_match(all_text, persona_ctx, llm)
+            report.voice_check_passed = voice_score >= 3.0
+            if voice_score < 3.0:
+                report.warnings.append(
+                    f"[{FailureCode.VOICE_MISMATCH}] Voice match score: {voice_score:.1f}/5.0"
+                )
+                logger.warning("Voice match: %.1f/5.0 — below threshold", voice_score)
+            else:
+                logger.info("Voice match: %.1f/5.0", voice_score)
+        except Exception as e:
+            logger.warning("Quality review/voice check failed: %s", e)
+            report.quality_review_passed = False
+            report.quality_review_issues = [f"Review failed: {type(e).__name__}"]
+
         # ── Build response ─────────────────────────────────────────────
         lines = []
 

{clawed-2.3.7 → clawed-2.3.8}/clawed/api/server.py

@@ -266,7 +266,9 @@ def create_app() -> FastAPI:
         filter_html = ""
         if all_subjects:
             opts = "".join(
-                f"<option value='{html_mod.escape(s)}' {'selected' if s == subject_filter else ''}>{html_mod.escape(s)}</option>"
+                f"<option value='{html_mod.escape(s)}'"
+                f" {'selected' if s == subject_filter else ''}"
+                f">{html_mod.escape(s)}</option>"
                 for s in all_subjects
             )
             filter_html += (
@@ -275,7 +277,9 @@ def create_app() -> FastAPI:
             )
         if all_grades:
             opts = "".join(
-                f"<option value='{html_mod.escape(g)}' {'selected' if g == grade_filter else ''}>{html_mod.escape(g)}</option>"
+                f"<option value='{html_mod.escape(g)}'"
+                f" {'selected' if g == grade_filter else ''}"
+                f">{html_mod.escape(g)}</option>"
                 for g in all_grades
             )
             filter_html += (

clawed-2.3.8/clawed/async_utils.py

@@ -0,0 +1,22 @@
+"""Shared async utility for running coroutines from sync or async contexts."""
+from __future__ import annotations
+
+import asyncio
+
+
+def run_async_safe(coro):
+    """Run an async coroutine, handling both sync and async calling contexts.
+
+    When called from inside a running event loop (e.g., agent_core tools),
+    uses a thread to avoid nested asyncio.run() errors.  When called from
+    plain sync code (CLI), uses asyncio.run() directly.
+    """
+    try:
+        asyncio.get_running_loop()
+        # We're inside an event loop — run in a worker thread
+        import concurrent.futures
+        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
+            return pool.submit(asyncio.run, coro).result(timeout=30)
+    except RuntimeError:
+        # No running loop — safe to use asyncio.run()
+        return asyncio.run(coro)

{clawed-2.3.7 → clawed-2.3.8}/clawed/demo/__init__.py

@@ -34,11 +34,12 @@ def load_all_demos() -> dict[str, dict[str, Any]]:
 def is_demo_mode(config: Any = None) -> bool:
     """Check whether the app should run in demo mode (no API key configured).
 
-    Args:
-        config: Optional AppConfig instance. When provided, the check uses
-            this config instead of loading global state. This allows
-            LLMClient to pass its own injected config.
+    Can be forced with CLAWED_DEMO=1 environment variable, which overrides
+    stored keys. This is useful for demo presentations and development.
     """
+    import os
+    if os.environ.get("CLAWED_DEMO", "").strip() in ("1", "true", "yes"):
+        return True
     from clawed.config import resolve_credentials
     provider, key = resolve_credentials(config)
     return provider is None

{clawed-2.3.7 → clawed-2.3.8}/clawed/export_docx.py

@@ -5,12 +5,12 @@ Generates full lesson plan documents and print-ready student worksheets.
 
 from __future__ import annotations
 
-import asyncio
 import logging
 from datetime import date
 from pathlib import Path
 from typing import TYPE_CHECKING, Any
 
+from clawed.async_utils import run_async_safe
 from clawed.export_theme import _resolve_output, get_color_theme
 
 if TYPE_CHECKING:
@@ -19,24 +19,6 @@ if TYPE_CHECKING:
 logger = logging.getLogger(__name__)
 
 
-def _run_async_safe(coro):
-    """Run an async coroutine, handling both sync and async calling contexts.
-
-    When called from inside a running event loop (e.g., agent_core tools),
-    uses a thread to avoid nested asyncio.run() errors.  When called from
-    plain sync code (CLI), uses asyncio.run() directly.
-    """
-    try:
-        asyncio.get_running_loop()
-        # We're inside an event loop — run in a worker thread
-        import concurrent.futures
-        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
-            return pool.submit(asyncio.run, coro).result(timeout=30)
-    except RuntimeError:
-        # No running loop — safe to use asyncio.run()
-        return asyncio.run(coro)
-
-
 # ── Image helpers (DOCX-specific) ─────────────────────────────────────
 
 
@@ -57,7 +39,7 @@ def _docx_add_image(
 
         from clawed.slide_images import fetch_slide_image
 
-        img_path = _run_async_safe(fetch_slide_image(topic, subject=subject))
+        img_path = run_async_safe(fetch_slide_image(topic, subject=subject))
         if img_path and img_path.exists():
             doc.add_picture(str(img_path), width=Inches(width_inches))
             # Center the image
@@ -97,7 +79,7 @@ def _docx_add_content_image(
 
         from clawed.slide_images import _extract_key_concepts, fetch_content_image
 
-        img_path = _run_async_safe(
+        img_path = run_async_safe(
             fetch_content_image(
                 content_text,
                 subject=subject,

{clawed-2.3.7 → clawed-2.3.8}/clawed/export_pptx.py

@@ -12,6 +12,7 @@ from datetime import date
 from pathlib import Path
 from typing import TYPE_CHECKING, Optional
 
+from clawed.async_utils import run_async_safe
 from clawed.export_theme import _hex_to_rgb, _resolve_output, get_color_theme
 
 if TYPE_CHECKING:
@@ -109,24 +110,6 @@ def _section_divider(prs, slide_num, text, theme, slide_w, slide_h):
 # ── Image fetching ────────────────────────────────────────────────────
 
 
-def _run_async_safe(coro):
-    """Run an async coroutine, handling both sync and async calling contexts.
-
-    When called from inside a running event loop (e.g., agent_core tools),
-    uses a thread to avoid nested asyncio.run() errors.  When called from
-    plain sync code (CLI), uses asyncio.run() directly.
-    """
-    try:
-        asyncio.get_running_loop()
-        # We're inside an event loop — run in a worker thread
-        import concurrent.futures
-        with concurrent.futures.ThreadPoolExecutor(max_workers=1) as pool:
-            return pool.submit(asyncio.run, coro).result(timeout=30)
-    except RuntimeError:
-        # No running loop — safe to use asyncio.run()
-        return asyncio.run(coro)
-
-
 def _try_fetch_images(topics: list[tuple[str, str]], subject: str) -> dict[str, Optional[Path]]:
     """Attempt to fetch images for multiple topics. Non-blocking, short timeout.
 
@@ -148,7 +131,7 @@ def _try_fetch_images(topics: list[tuple[str, str]], subject: str) -> dict[str,
                 results[key] = None
 
     try:
-        _run_async_safe(_fetch_all())
+        run_async_safe(_fetch_all())
     except Exception as e:
         logger.debug("Image fetching failed: %s", e)
 
@@ -200,7 +183,7 @@ def _try_fetch_content_images(
                 results[key] = None
 
     try:
-        _run_async_safe(_fetch_all())
+        run_async_safe(_fetch_all())
     except Exception as e:
         logger.debug("Content image fetching failed: %s", e)
 

clawed-2.3.8/clawed/failure_codes.py

@@ -0,0 +1,22 @@
+"""NLAH failure taxonomy — structured failure codes for generation pipeline."""
+from __future__ import annotations
+
+from enum import Enum
+
+
+class FailureCode(str, Enum):
+    """Machine-parseable failure codes per NLAH Section 6."""
+
+    NO_PERSONA = "NO_PERSONA"
+    SCHEMA_ERROR = "SCHEMA_ERROR"
+    TOPIC_DRIFT = "TOPIC_DRIFT"
+    DEMO_FIXTURE = "DEMO_FIXTURE"
+    EXPORT_INCOMPLETE = "EXPORT_INCOMPLETE"
+    EXPORT_ERROR = "EXPORT_ERROR"
+    REVIEW_FAILED = "REVIEW_FAILED"
+    CONTEXT_EXCEEDED = "CONTEXT_EXCEEDED"
+    API_FAILURE = "API_FAILURE"
+    VOICE_MISMATCH = "VOICE_MISMATCH"
+    PERSONA_PARSE_ERROR = "PERSONA_PARSE_ERROR"
+    KB_SEARCH_FAILED = "KB_SEARCH_FAILED"
+    ASSET_SEARCH_FAILED = "ASSET_SEARCH_FAILED"

{clawed-2.3.7 → clawed-2.3.8}/clawed/generation.py

@@ -624,12 +624,11 @@ async def handle_connect_local(
                     progress_callback=_progress,
                 )
                 if docs:
-                    import asyncio
-
+                    from clawed.async_utils import run_async_safe
                     from clawed.persona import extract_persona
 
                     persona_cfg = AppConfig.load()
-                    persona = asyncio.run(extract_persona(docs, persona_cfg))
+                    persona = run_async_safe(extract_persona(docs, persona_cfg))
                     session.persona = persona
                     session.save()
                     notify_callback(

{clawed-2.3.7 → clawed-2.3.8}/clawed/handlers/onboard.py

@@ -5,12 +5,15 @@ Extracted from tg.py lines 1156-1263.
 """
 from __future__ import annotations
 
+import logging
 import re
 from enum import Enum
 
 from clawed.gateway_response import GatewayResponse
 from clawed.models import AppConfig, TeacherPersona, TeacherProfile
 
+logger = logging.getLogger(__name__)
+
 
 class OnboardState(Enum):
     ASK_SUBJECT = "ask_subject"
@@ -84,7 +87,7 @@ class OnboardHandler:
 
         if current == OnboardState.ASK_SUBJECT:
             grade, subject = _parse_grade_and_subject(text)
-            state["subject"] = subject if subject else text.strip().title()
+            state["subject"] = (subject if subject else text.strip().title())[:100]
             if grade:
                 state["grade"] = grade
                 state["step"] = OnboardState.ASK_NAME
@@ -103,14 +106,21 @@ class OnboardHandler:
             elif re.search(r"(?:kindergarten|kinder|pre-?k)", text, re.IGNORECASE):
                 state["grade"] = "K"
             else:
-                state["grade"] = text.strip()
+                # Invalid grade — re-prompt
+                return GatewayResponse(
+                    text="I didn't catch that — what grade level? (K, 1-12, or Pre-K)"
+                )
             state["step"] = OnboardState.ASK_NAME
             return GatewayResponse(
                 text=f"Grade {state['grade']} — got it.\n\nWhat's your name?"
             )
 
         if current == OnboardState.ASK_NAME:
-            state["name"] = text.strip()
+            name = text.strip()[:100]
+            name = re.sub(r'[^\w\s\'-]', '', name).strip()
+            if not name:
+                return GatewayResponse(text="I need a name to personalize your lessons. What should I call you?")
+            state["name"] = name
             return self._complete_onboarding(teacher_id)
 
         return GatewayResponse(text="Something went wrong with setup. Try /start again.")
@@ -132,8 +142,8 @@ class OnboardHandler:
             from clawed.workspace import init_workspace
             persona = TeacherPersona(name=state["name"], subject_area=state["subject"])
             init_workspace(persona, config)
-        except Exception:
-            pass
+        except Exception as e:
+            logger.warning("Workspace init failed during onboarding: %s", e)
 
         del self._state[teacher_id]
 

{clawed-2.3.7 → clawed-2.3.8}/clawed/ingestor.py

@@ -457,7 +457,7 @@ def _extract_xlsx(path: Path) -> str:
             rows.append(f"[Sheet: {sheet}]")
             for i, row in enumerate(ws.iter_rows(values_only=True)):
                 if i >= 5000:
-                    rows.append(f"... (truncated at 5000 rows)")
+                    rows.append("... (truncated at 5000 rows)")
                     break
                 cells = [str(c) if c is not None else "" for c in row]
                 if any(cells):

{clawed-2.3.7 → clawed-2.3.8}/clawed/lesson.py

@@ -22,7 +22,9 @@ def _build_system_prompt(
     persona_context = persona.to_prompt_context() if persona else ""
     soul_context = ""
     try:
-        soul_path = Path.home() / ".eduagent" / "workspace" / "SOUL.md"
+        import os
+        data_dir = os.environ.get("EDUAGENT_DATA_DIR", str(Path.home() / ".eduagent"))
+        soul_path = Path(data_dir) / "workspace" / "SOUL.md"
         if soul_path.exists():
             soul_context = soul_path.read_text(encoding="utf-8")[:2000]
     except Exception:

{clawed-2.3.7 → clawed-2.3.8}/clawed/llm.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import json
+import logging
 import os
 from typing import Any, Optional, Type
 
@@ -11,6 +12,8 @@ from pydantic import BaseModel, ValidationError
 
 from clawed.models import AppConfig, LLMProvider
 
+logger = logging.getLogger(__name__)
+
 
 class LLMClient:
     """Unified async LLM client for all supported backends."""
@@ -414,35 +417,39 @@ class LLMClient:
         """Self-review a lesson package against observation-ready standards.
 
         Returns a dict with 'passed' (bool) and 'issues' (list of strings).
-        If issues are found, the caller should attempt to fix them.
+        Fails closed: any exception returns passed=False (NLAH Section 3, Stage 4).
         """
-        prompt = (
-            f"Review this lesson package against observation-ready quality standards.\n\n"
-            f"Lesson:\n{lesson_json[:3000]}\n\n"
-            f"Package status: standards={'yes' if standards_present else 'MISSING'}, "
-            f"handout={'yes' if has_handout else 'MISSING'}, "
-            f"slideshow={'yes' if has_slideshow else 'MISSING'}\n\n"
-            "Check these standards:\n"
-            "1. Do all section times add up to a full class period (42-45 min)?\n"
-            "2. Are specific standards codes listed (not just 'aligned to standards')?\n"
-            "3. Is vocabulary defined for all content-specific terms?\n"
-            "4. Are there checks for understanding every 7-10 minutes?\n"
-            "5. Are all referenced materials self-contained (no phantom handouts)?\n"
-            "6. Is the Do Now completable in 5 minutes?\n"
-            "7. Are transitions scripted between sections?\n\n"
-            'Return JSON: {"passed": true/false, "issues": ["issue 1", "issue 2"]}'
-        )
-        raw = await self.generate(prompt, temperature=0.2, max_tokens=1000)
         try:
+            prompt = (
+                f"Review this lesson package against observation-ready quality standards.\n\n"
+                f"Lesson:\n{lesson_json[:3000]}\n\n"
+                f"Package status: standards={'yes' if standards_present else 'MISSING'}, "
+                f"handout={'yes' if has_handout else 'MISSING'}, "
+                f"slideshow={'yes' if has_slideshow else 'MISSING'}\n\n"
+                "Check these standards:\n"
+                "1. Do all section times add up to a full class period (42-45 min)?\n"
+                "2. Are specific standards codes listed (not just 'aligned to standards')?\n"
+                "3. Is vocabulary defined for all content-specific terms?\n"
+                "4. Are there checks for understanding every 7-10 minutes?\n"
+                "5. Are all referenced materials self-contained (no phantom handouts)?\n"
+                "6. Is the Do Now completable in 5 minutes?\n"
+                "7. Are transitions scripted between sections?\n\n"
+                'Return JSON: {"passed": true/false, "issues": ["issue 1", "issue 2"]}'
+            )
+            raw = await self.generate(prompt, temperature=0.2, max_tokens=1000)
             cleaned = raw.strip()
             if cleaned.startswith("```"):
                 lines = cleaned.split("\n")[1:]
                 if lines and lines[-1].strip() == "```":
                     lines = lines[:-1]
                 cleaned = "\n".join(lines)
-            return json.loads(cleaned)
-        except Exception:
-            return {"passed": False, "issues": ["Quality review could not parse LLM response — review skipped"]}
+            result = json.loads(cleaned)
+            if "passed" not in result:
+                return {"passed": False, "issues": ["LLM response missing 'passed' field"]}
+            return result
+        except Exception as e:
+            logger.warning("REVIEW_FAILED: %s", e)
+            return {"passed": False, "issues": [f"Quality review failed: {type(e).__name__}"]}
 
     # ── Anthropic ────────────────────────────────────────────────────────
 

{clawed-2.3.7 → clawed-2.3.8}/clawed/quality.py

@@ -131,3 +131,54 @@ class LessonQualityScore:
         if materials.assessment_questions:
             parts.append(f"Assessment: {len(materials.assessment_questions)} questions")
         return "\n".join(parts)
+
+
+async def score_voice_match(
+    lesson_text: str,
+    persona_context: str,
+    llm_client: Any = None,
+) -> float:
+    """Score how well lesson text matches the teacher's voice (1.0-5.0).
+
+    Uses an LLM to compare the lesson language, tone, scaffolding style,
+    and pedagogical patterns against the teacher persona.
+
+    Returns 3.0 (neutral) if scoring fails — NLAH: do not block delivery.
+    """
+    if not persona_context or not lesson_text:
+        return 3.0
+
+    prompt = (
+        "Rate how well this lesson matches the teacher's established voice "
+        "and teaching style on a scale of 1.0 to 5.0.\n\n"
+        "Teacher persona:\n"
+        f"{persona_context[:1500]}\n\n"
+        "Lesson excerpt:\n"
+        f"{lesson_text[:2000]}\n\n"
+        "Score criteria:\n"
+        "5.0 = Sounds exactly like this teacher wrote it\n"
+        "4.0 = Captures most of their style and patterns\n"
+        "3.0 = Generic but acceptable\n"
+        "2.0 = Noticeably different from their voice\n"
+        "1.0 = Completely wrong tone/style\n\n"
+        'Return ONLY a JSON object: {"score": 4.2, "reason": "brief explanation"}'
+    )
+
+    if llm_client is None:
+        return 3.0  # No LLM available, neutral score
+
+    try:
+        import json as _json
+
+        raw = await llm_client.generate(prompt, temperature=0.2, max_tokens=200)
+        cleaned = raw.strip()
+        if cleaned.startswith("```"):
+            lines = cleaned.split("\n")[1:]
+            if lines and lines[-1].strip() == "```":
+                lines = lines[:-1]
+            cleaned = "\n".join(lines)
+        result = _json.loads(cleaned)
+        score = float(result.get("score", 3.0))
+        return max(1.0, min(5.0, score))  # Clamp to valid range
+    except Exception:
+        return 3.0  # Fail neutral — NLAH: do not block delivery