fix(tools): MP 工具改服务端限量查询，止血 IP 被封根因

mp_search_summary 之前不传分页参数，mp-api 默认 chunk_size=1000 且
list(docs) 自动翻完所有页，limit 只做客户端切片 —— 每次搜索都整库级下载，
被 MP 判 abusive traffic 封 host IP/ASN。改为 search(num_chunks=1,
chunk_size=limit, ...) 服务端单页限量。mp_get_entries 的 limit 同为"只裁剪
保存、不减流量"的假参数，但 get_entries_in_chemsys 天然全量(相图用途)无法
限量，只在 description 点明开销。测试加断言锁定 num_chunks/chunk_size 已传。

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/test_secret_host_tools.py

@@ -75,9 +75,11 @@ class TestMaterialsProjectHostTools(unittest.TestCase):
             formula_pretty = "Ca3SiO5"
             energy_above_hull = 0.0123
 
+        captured = {}
+
         class FakeSummary:
             def search(self, **kwargs):
-                self.kwargs = kwargs
+                captured.update(kwargs)
                 return [FakeDoc()]
 
         class FakeMaterials:
@@ -110,6 +112,10 @@ class TestMaterialsProjectHostTools(unittest.TestCase):
         self.assertEqual(data[0]["formula_pretty"], "Ca3SiO5")
         self.assertEqual(data[0]["energy_above_hull"], 0.0123)
         self.assertNotIn("host-secret", out)
+        # Server-side limiting: a single page of `limit` records, never the
+        # default chunk_size=1000 full-database pagination that gets the IP banned.
+        self.assertEqual(captured["num_chunks"], 1)
+        self.assertEqual(captured["chunk_size"], 2)
 
 
 if __name__ == "__main__":

tools/materials_project.py

@@ -70,8 +70,8 @@ class MaterialsProjectSearchSummaryTool(Tool):
             "formula": {"type": "string", "description": "Optional formula such as Ca3SiO5."},
             "material_ids": {"type": "array", "items": {"type": "string"}, "description": "Optional Materials Project ids such as mp-123."},
             "elements": {"type": "array", "items": {"type": "string"}, "description": "Optional element symbols for a chemical system search."},
-            "fields": {"type": "array", "items": {"type": "string"}, "description": "Fields to return; defaults to material_id/formula/symmetry/energy_above_hull."},
-            "limit": {"type": "integer", "default": 10, "description": "Maximum records returned, 1-50."},
+            "fields": {"type": "array", "items": {"type": "string"}, "description": "Fields to return; defaults to material_id/formula_pretty/symmetry/energy_above_hull. Do NOT put the search formula here — use formula_pretty (not 'formula'). Common fields: formula_pretty, symmetry, energy_above_hull, band_gap, density, volume, is_stable, structure."},
+            "limit": {"type": "integer", "default": 10, "description": "Maximum records returned, 1-50. Server-side bounded — keep small to stay within MP fair-use."},
         },
     }
 
@@ -96,7 +96,12 @@ class MaterialsProjectSearchSummaryTool(Tool):
             return "[Error] formula / material_ids / elements 至少传一个"
         try:
             with _mpr() as mpr:
-                docs = mpr.materials.summary.search(**kwargs)
+                # num_chunks=1 + chunk_size=limit 让服务端单次只回 limit 条、不翻页。
+                # 否则 mp-api 默认 chunk_size=1000 且自动拉完所有页 —— 整库级下载,
+                # 会被 MP 判定为 abusive traffic 并封 IP/ASN。
+                docs = mpr.materials.summary.search(
+                    num_chunks=1, chunk_size=limit, **kwargs
+                )
         except Exception as e:
             return f"[Error] mp_search_summary failed: {type(e).__name__}: {e}"
         plain = [_to_plain(d) for d in list(docs)[:limit]]
@@ -149,13 +154,14 @@ class MaterialsProjectGetEntriesTool(Tool):
     name = "mp_get_entries"
     description = (
         "Fetch Materials Project computed entries for a chemical system and save trimmed JSON to task_dir/materials/. "
+        "Downloads the FULL chemical system (all sub-systems) — volume grows fast with element count; call sparingly and reuse the saved file rather than re-querying."
     )
     parameters = {
         "type": "object",
         "properties": {
-            "elements": {"type": "array", "items": {"type": "string"}, "description": "Chemical system elements, e.g. ['Ca','Si','O','H']."},
+            "elements": {"type": "array", "items": {"type": "string"}, "description": "Chemical system elements, e.g. ['Ca','Si','O','H']. More elements = much heavier download."},
             "filename": {"type": "string", "description": "Optional JSON filename. Defaults to mp_entries_<chemsys>.json."},
-            "limit": {"type": "integer", "default": 200, "description": "Maximum entries saved, 1-1000."},
+            "limit": {"type": "integer", "default": 200, "description": "Max entries SAVED to disk, 1-1000. NOTE: this only trims the saved JSON — the full chemsys is still fetched from MP. It does not reduce network traffic."},
         },
         "required": ["elements"],
     }