fix: resolve all 173 mypy type errors across 8 packages

packages/arxiv-mcp/src/arxiv_mcp/server.py

@@ -76,11 +76,19 @@ def parse_arxiv_entry(entry: ET.Element) -> dict[str, Any]:
     """
     # Extract basic metadata
     title = entry.find("atom:title", ARXIV_NAMESPACE)
-    title_text = title.text.strip().replace("\n", " ") if title is not None else "Unknown"
+    title_text = (
+        title.text.strip().replace("\n", " ")
+        if title is not None and title.text is not None
+        else "Unknown"
+    )
 
     # Extract ArXiv ID from the entry ID
     entry_id = entry.find("atom:id", ARXIV_NAMESPACE)
-    arxiv_id = entry_id.text.split("/abs/")[-1] if entry_id is not None else "Unknown"
+    arxiv_id = (
+        entry_id.text.split("/abs/")[-1]
+        if entry_id is not None and entry_id.text is not None
+        else "Unknown"
+    )
 
     # Extract authors
     authors = []
@@ -91,7 +99,11 @@ def parse_arxiv_entry(entry: ET.Element) -> dict[str, Any]:
 
     # Extract summary (abstract)
     summary = entry.find("atom:summary", ARXIV_NAMESPACE)
-    summary_text = summary.text.strip().replace("\n", " ") if summary is not None else ""
+    summary_text = (
+        summary.text.strip().replace("\n", " ")
+        if summary is not None and summary.text is not None
+        else ""
+    )
 
     # Extract published date
     published = entry.find("atom:published", ARXIV_NAMESPACE)
@@ -186,7 +198,8 @@ async def call_openai_api(prompt: str, model: str = "gpt-4") -> str | None:
             response = await client.post(url, headers=headers, json=payload, timeout=60.0)
             response.raise_for_status()
             result = response.json()
-            return result["choices"][0]["message"]["content"]
+            content: str | None = result["choices"][0]["message"]["content"]
+            return content
         except httpx.HTTPError as e:
             logger.error(f"OpenAI API error: {e}")
             return None
@@ -229,7 +242,8 @@ async def call_anthropic_api(prompt: str, model: str = "claude-3-5-sonnet-202410
             response = await client.post(url, headers=headers, json=payload, timeout=60.0)
             response.raise_for_status()
             result = response.json()
-            return result["content"][0]["text"]
+            text: str | None = result["content"][0]["text"]
+            return text
         except httpx.HTTPError as e:
             logger.error(f"Anthropic API error: {e}")
             return None
@@ -390,7 +404,7 @@ async def summarize_paper_with_llm(
         A concise summary of the paper generated by the LLM
     """
     # First, fetch the paper details
-    paper_info = await get_arxiv_paper(arxiv_id)
+    paper_info: str = await get_arxiv_paper(arxiv_id)
 
     if paper_info.startswith("Error:") or paper_info.startswith("Paper not found:"):
         return paper_info
@@ -460,7 +474,7 @@ async def search_and_summarize(
     max_papers = min(max_papers, 5)
 
     # First, search for papers
-    search_results = await search_arxiv(query, max_results=max_papers)
+    search_results: str = await search_arxiv(query, max_results=max_papers)
 
     if search_results.startswith("Error:") or search_results.startswith("No papers found"):
         return search_results
@@ -481,7 +495,7 @@ async def search_and_summarize(
 
     for i, arxiv_id in enumerate(arxiv_ids, 1):
         logger.info(f"Summarizing paper {i}/{len(arxiv_ids)}: {arxiv_id}")
-        summary = await summarize_paper_with_llm(arxiv_id, llm_provider)
+        summary: str = await summarize_paper_with_llm(arxiv_id, llm_provider)
         summaries.append(f"\n{i}. {summary}")
         summaries.append("=" * 80)
 

packages/bgs-mcp/src/bgs_mcp/api.py

@@ -205,7 +205,7 @@ async def list_commodities(
     else:
         commodities = await client.get_commodities()
 
-    categories = None
+    categories: dict[str, list[str]] | None = None
     if categorize:
         categories = {
             "battery": [],
@@ -443,17 +443,17 @@ async def get_time_series(
 
     # Aggregate by year if no country specified
     if not country:
-        year_totals = {}
+        year_totals: dict[int, float] = {}
         units = None
         for r in records:
             if r.year and r.quantity is not None:
                 if r.year not in year_totals:
-                    year_totals[r.year] = 0
+                    year_totals[r.year] = 0.0
                 year_totals[r.year] += r.quantity
                 units = r.units
 
         data = []
-        prev_qty = None
+        prev_qty: float | None = None
         for year in sorted(year_totals.keys()):
             qty = year_totals[year]
             yoy = None
@@ -474,16 +474,16 @@ async def get_time_series(
         units = records[0].units if records else None
 
         data = []
-        prev_qty = None
+        prev_qty_r: float | None = None
         for r in records:
             if r.year and r.quantity is not None:
                 yoy = None
-                if prev_qty and prev_qty > 0:
-                    yoy = round(((r.quantity - prev_qty) / prev_qty) * 100, 2)
+                if prev_qty_r and prev_qty_r > 0:
+                    yoy = round(((r.quantity - prev_qty_r) / prev_qty_r) * 100, 2)
                 data.append(
                     TimeSeriesPoint(year=r.year, quantity=r.quantity, yoy_change_percent=yoy)
                 )
-                prev_qty = r.quantity
+                prev_qty_r = r.quantity
 
         return TimeSeriesResponse(
             commodity=commodity,
@@ -554,7 +554,7 @@ async def get_country_profile(
     client = get_client()
 
     country_iso = None
-    country_name = country
+    country_name: str | None = country
     if len(country) <= 3:
         country_iso = country
         country_name = None
@@ -576,7 +576,7 @@ async def get_country_profile(
         year = max(available_years)
 
     # Aggregate by commodity
-    commodity_data = {}
+    commodity_data: dict[str, dict[str, Any]] = {}
     for r in records:
         if r.year != year:
             continue
@@ -585,19 +585,19 @@ async def get_country_profile(
 
         key = r.commodity
         if key not in commodity_data:
-            commodity_data[key] = {"commodity": key, "quantity": 0, "units": r.units}
+            commodity_data[key] = {"commodity": key, "quantity": 0.0, "units": r.units}
         commodity_data[key]["quantity"] += r.quantity
 
     # Sort by quantity
     commodities = sorted(
         commodity_data.values(),
-        key=lambda x: x["quantity"],
+        key=lambda x: float(x["quantity"]),
         reverse=True,
     )
 
     return CountryProfile(
         country=actual_country,
-        year=year,
+        year=year if year is not None else 0,
         statistic_type=statistic_type,
         commodities=commodities,
     )

packages/bgs-mcp/src/bgs_mcp/bgs_client.py

@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-from typing import Any
+from typing import Any, cast
 
 import httpx
 from pydantic import BaseModel
@@ -104,7 +104,7 @@ async def _request(
                 headers={"Accept": "application/json"},
             )
             response.raise_for_status()
-            return response.json()
+            return cast(dict[str, Any], response.json())
 
     def _parse_records(self, data: dict[str, Any]) -> list[MineralRecord]:
         """Parse API response into MineralRecord objects."""
@@ -206,7 +206,7 @@ async def search_production(
                 params["country_iso3_code"] = country_iso.upper()
 
         # Fetch data
-        all_records = []
+        all_records: list[MineralRecord] = []
         offset = 0
 
         while len(all_records) < limit:
@@ -303,7 +303,7 @@ async def get_commodity_by_country(
             year = max(r.year for r in records if r.year)
 
         # Filter to target year and aggregate by country
-        country_totals = {}
+        country_totals: dict[str, dict[str, Any]] = {}
 
         for record in records:
             if record.year != year:
@@ -316,7 +316,7 @@ async def get_commodity_by_country(
                 country_totals[country] = {
                     "country": country,
                     "country_iso3": record.country_iso3,
-                    "quantity": 0,
+                    "quantity": 0.0,
                     "units": record.units,
                     "year": year,
                 }
@@ -325,7 +325,7 @@ async def get_commodity_by_country(
         # Sort by quantity descending
         ranked = sorted(
             country_totals.values(),
-            key=lambda x: x["quantity"],
+            key=lambda x: float(x["quantity"]),
             reverse=True,
         )
 

packages/bgs-mcp/src/bgs_mcp/server.py

@@ -287,12 +287,12 @@ async def get_time_series(
 
     # If no country specified, aggregate by year
     if not country:
-        year_totals = {}
+        year_totals: dict[int, float] = {}
         units = None
         for r in records:
             if r.year and r.quantity is not None:
                 if r.year not in year_totals:
-                    year_totals[r.year] = 0
+                    year_totals[r.year] = 0.0
                 year_totals[r.year] += r.quantity
                 units = r.units
 
@@ -323,16 +323,16 @@ async def get_time_series(
         output += "| Year | Quantity | YoY Change |\n"
         output += "|------|----------|------------|\n"
 
-        prev_qty = None
+        prev_qty_r: float | None = None
         for r in records:
             if r.year and r.quantity is not None:
-                if prev_qty and prev_qty > 0:
-                    change = ((r.quantity - prev_qty) / prev_qty) * 100
+                if prev_qty_r and prev_qty_r > 0:
+                    change = ((r.quantity - prev_qty_r) / prev_qty_r) * 100
                     change_str = f"{change:+.1f}%"
                 else:
                     change_str = "-"
                 output += f"| {r.year} | {r.quantity:,.1f} | {change_str} |\n"
-                prev_qty = r.quantity
+                prev_qty_r = r.quantity
 
     return output
 
@@ -431,7 +431,7 @@ async def get_country_profile(
     client = get_client()
 
     country_iso = None
-    country_name = country
+    country_name: str | None = country
     if len(country) <= 3:
         country_iso = country
         country_name = None
@@ -454,7 +454,7 @@ async def get_country_profile(
         year = max(available_years)
 
     # Filter to target year and aggregate by commodity
-    commodity_data = {}
+    commodity_data: dict[str, dict[str, float | str | None]] = {}
     for r in records:
         if r.year != year:
             continue
@@ -463,8 +463,8 @@ async def get_country_profile(
 
         key = r.commodity
         if key not in commodity_data:
-            commodity_data[key] = {"quantity": 0, "units": r.units}
-        commodity_data[key]["quantity"] += r.quantity
+            commodity_data[key] = {"quantity": 0.0, "units": r.units}
+        commodity_data[key]["quantity"] = float(commodity_data[key]["quantity"] or 0) + r.quantity
 
     output = f"**{actual_country} - {statistic_type} Profile ({year})**\n\n"
     output += f"Commodities: {len(commodity_data)}\n\n"
@@ -475,7 +475,7 @@ async def get_country_profile(
     # Sort by quantity descending
     sorted_commodities = sorted(
         commodity_data.items(),
-        key=lambda x: x[1]["quantity"],
+        key=lambda x: float(x[1]["quantity"] or 0),
         reverse=True,
     )
 

packages/claimm-mcp/src/claimm_mcp/config.py

@@ -100,5 +100,5 @@ def get_settings() -> Settings:
     """Get or create the settings instance."""
     global _settings
     if _settings is None:
-        _settings = Settings()
+        _settings = Settings()  # type: ignore[call-arg]
     return _settings

packages/claimm-mcp/src/claimm_mcp/edx_client.py

@@ -83,7 +83,8 @@ async def _request(
                 error = result.get("error", {})
                 raise Exception(f"EDX API error: {error}")
 
-            return result.get("result", {})
+            api_result: dict[str, Any] = result.get("result", {})
+            return api_result
 
     async def search_resources(
         self,

packages/claimm-mcp/src/claimm_mcp/header_detector.py

@@ -150,7 +150,7 @@ def _detect_delimiter(self, line: str) -> str:
         """Auto-detect CSV delimiter."""
         delimiters = [",", "\t", ";", "|"]
         counts = {d: line.count(d) for d in delimiters}
-        return max(counts, key=counts.get) if max(counts.values()) > 0 else ","
+        return max(counts, key=lambda d: counts[d]) if max(counts.values()) > 0 else ","
 
     def _detect_column_types(
         self,

packages/claimm-mcp/src/claimm_mcp/server.py

@@ -162,7 +162,7 @@ async def detect_dataset_schemas(
                 output += "\n"
 
                 # Show column types summary
-                types = {}
+                types: dict[str, int] = {}
                 for col in result.get("column_types", []):
                     t = col.get("type", "unknown")
                     types[t] = types.get(t, 0) + 1
@@ -428,9 +428,9 @@ async def ask_about_data(
     if dataset_id:
         submission = await edx.get_submission(dataset_id)
         # Use the first resource as context if available
-        resource = submission.resources[0] if submission.resources else None
-        if resource:
-            return await llm.answer_about_resource(resource, submission, question)
+        first_resource = submission.resources[0] if submission.resources else None
+        if first_resource:
+            return await llm.answer_about_resource(first_resource, submission, question)
         else:
             # Answer based on submission only
             try:

packages/cmm-data/src/cmm_data/__init__.py

@@ -88,7 +88,7 @@ def load_ore_deposits(table: str = "all"):
     from .loaders.usgs_ore import USGSOreDepositsLoader
 
     loader = USGSOreDepositsLoader()
-    return loader.load(table)
+    return loader.load(table=table)
 
 
 def search_documents(query: str, **kwargs):

packages/cmm-data/src/cmm_data/catalog.py

@@ -157,7 +157,7 @@ def search_all_datasets(query: str, datasets: list[str] | None = None) -> pd.Dat
 
     if "usgs_commodity" in datasets:
         try:
-            loader = USGSCommodityLoader()
+            USGSCommodityLoader()  # Validate loader availability
             # Search commodity names
             for code, name in COMMODITY_NAMES.items():
                 if query.lower() in name.lower() or query.lower() in code.lower():
@@ -175,8 +175,8 @@ def search_all_datasets(query: str, datasets: list[str] | None = None) -> pd.Dat
 
     if "osti" in datasets:
         try:
-            loader = OSTIDocumentsLoader()
-            docs = loader.search_documents(query, limit=20)
+            osti_loader = OSTIDocumentsLoader()
+            docs = osti_loader.search_documents(query, limit=20)
             for _, row in docs.iterrows():
                 results.append(
                     {
@@ -192,8 +192,8 @@ def search_all_datasets(query: str, datasets: list[str] | None = None) -> pd.Dat
 
     if "preprocessed" in datasets:
         try:
-            loader = PreprocessedCorpusLoader()
-            docs = loader.search(query, limit=20)
+            preprocessed_loader = PreprocessedCorpusLoader()
+            docs = preprocessed_loader.search(query, limit=20)
             for _, row in docs.iterrows():
                 results.append(
                     {