feat(expert-distiller): cross-validated-search maturity 72/100

MATURITY_REPORT.md

@@ -0,0 +1,56 @@
+# Maturity Report: Cross-Validated Search
+
+**Domain**: cross-validated-search
+**Iteration**: 2 / 10
+**Status**: running
+
+## Score Summary
+
+| Axis | Score | Max |
+|------|-------|-----|
+| Breadth | 16 | 25 |
+| Depth | 18 | 25 |
+| Thickness | 18 | 25 |
+| Effectiveness | 17 | 25 |
+| **Total** | **69** | **100** |
+
+## Expert Council
+
+| Expert | Role | Confidence | Weight |
+|--------|------|-----------|--------|
+| andreas-hanselowski | skeptic | high | 0.2 |
+| isabelle-augenstein | advocate | high | 0.2 |
+| pepa-atanasova | reviewer | high | 0.24 |
+| preslav-nakov | reviewer | high | 0.24 |
+| reihaneh-rabbany | chair | high | 0.3 |
+
+## Iteration History
+
+| Iteration | Total | Breadth | Depth | Thickness | Effectiveness | Action |
+|-----------|-------|---------|-------|-----------|--------------|--------|
+| 1 | 69 | 16 | 18 | 18 | 17 | Initial scoring with 5-expert council |
+
+## Coverage Gaps
+
+- No dedicated MCP protocol specialist; Hanselowski covers agentic workflows but not the MCP specification itself
+- Information extraction from web pages (axis 7) still lacks a dedicated expert; browse_page uses regex/BeautifulSoup without main content detection
+- Multilingual concerns are mentioned by Nakov and Atanasova but not deeply operationalized in any expert's reasoning kernel
+- Production deployment, latency, and adversarial robustness remain acknowledged blind spots without deep coverage
+- Adversarial robustness of credibility scoring and search result manipulation is a cross-cutting blind spot
+- Nakov's source_confidence is 0.75 with only 2 source refs, limiting verifiable source alignment for his deeper playbook entries
+- Augenstein's quote bank has 2 entries, one of which is a paraphrase -- still below 'authoritative reference' standard
+- No profile reaches 23-25 (authoritative standalone reference work level); all remain in the 17-19 range
+- Blind spots are self-identified but not externally validated against the actual project codebase
+- Atanasova's playbook entries (4) are strong but her canonical works remain at 3, limiting depth of source-backed claims
+
+## Recommendations
+
+- CRITICAL: Fix 3 failing tests before further development. The semicolon claim decomposition regex needs adjustment, and Unicode word boundaries for accented characters need \b replaced with Unicode-aware alternatives or lookaround assertions.
+- BREADTH: Add an MCP protocol specialist or expand Hanselowski's profile to cover MCP tool specification patterns, sampling, and resource management in depth.
+- DEPTH: Augenstein's quote bank still needs more direct quotes (currently 2, one paraphrase). Nakov's source confidence is limited by only 2 source refs. Consider finding primary source materials for additional canonical works.
+- EFFECTIVENESS: Sub-claim verification should issue targeted sub-queries rather than reusing the parent claim's source pool. This is the single highest-impact improvement for effectiveness.
+- EFFECTIVENESS: Replace regex \b word boundaries with Unicode-aware alternatives for multilingual conflict detection (use \W|^|$ lookarounds or regex with re.UNICODE flag).
+- THICKNESS: Add mypy/pyright type checking and ruff linting to CI to prevent regressions.
+- EFFECTIVENESS: Consider simple baselines for verification (per Rabbany's playbook): majority-vote heuristic, keyword-count heuristic, and compare against the current weighted overlap heuristic to establish that the added complexity is justified.
+
+Generated by Expert Distiller at 2026-04-17T02:57:11Z

cross_validated_search/__init__.py

@@ -5,6 +5,7 @@
     CrossValidatedSearcher,
     EvidenceReportResult,
     Source,
+    SubClaimResult,
     UltimateSearcher,
     VerificationResult,
 )
@@ -17,6 +18,7 @@
     "EvidenceReportResult",
     "SearxngProvider",
     "Source",
+    "SubClaimResult",
     "UltimateSearcher",
     "VerificationResult",
 ]