v1.1.1: 升级 AI 续聘建议为 LLM + Playwright RPA 采集 SmartCampus 平台材料

新增：

- scripts/service/platform_collector.py：Playwright RPA（持久化 browser profile 复用登录态），抓合同表（姓名/单位/到期日/剩余天数/预警/建议），逐行点"续聘材料包"按钮下载 PDF/Word 到 data/materials/

- scripts/service/material_parser.py：PDF/Word 文本提取（pypdf → pdf2image + pytesseract OCR 扫描件兜底；python-docx 抽 docx；antiword 兜底 .doc）

- scripts/service/llm_decision.py：AI 决策统一入口（优先调宿主 jiangchang_skill_core.llm_manager.call_llm_json；本地降级到规则版；输出 decision_source ∈ {llm, rules, rules_no_materials}）

- scripts/db/migrations.py：idempotent ALTER/CREATE，建 materials_content 表 + renewal_tasks 加 decision_source / materials_summary 列

修改：

- scripts/db/connection.py：init_db() 自动调 apply_all() 应用 v1.1.1 迁移

- scripts/db/repository.py：加 insert_material / list_materials_for_contract / list_pending_materials / summarize_materials / get_task_full

- scripts/service/task_generator.py：AI 决策统一改为调 llm_decision.decide_suggestion(materials_summary=...)；reason 文案加「已结合材料包（X 份文件，Y 字 / 待解析）综合判断：...」

- scripts/service/html_renderer.py：ctx 加 DECISION_SOURCE（LLM 推理 / 本地规则 / 本地规则（无材料））+ MATERIALS_COUNT

- scripts/templates/task.html.j2：footer 加 "AI 建议来源：{{DECISION_SOURCE}} · 已审阅材料 {{MATERIALS_COUNT}} 份"

- scripts/cli/app.py：新增 2 个 CLI（collect-from-platform / parse-materials）；现有 9 个不动

- requirements.txt：playwright / pypdf / python-docx / pdf2image / pytesseract / Pillow

- .env.example：SMARTCAMPUS_BASE_URL / SMARTCAMPUS_PROFILE_DIR / MATERIALS_DOWNLOAD_DIR / OCR_LANG / LLM_MODEL_NAME / LLM_API_BASE_URL / LLM_API_TOKEN

- SKILL.md：version 1.1.0 → 1.1.1；用户意图路由表加 2 行；新增「数据来源」段（三层来源 + AI 决策透明化）

- README.md：使用前要准备 → 加 playwright install chromium 步骤

- CHANGELOG.md：v1.1.1 条目

- scripts/util/constants.py：SKILL_VERSION 1.1.0 → 1.1.1

本地验证：11 个 CLI 全部 exit 0；decision_source=rules；reason 含「已结合材料包（6 份文件，待解析）」；HTML footer 含决策来源。

Co-Authored-By: Claude Code <noreply@anthropic.com>

scripts/db/repository.py

@@ -126,13 +126,16 @@ def insert_task(
     reason: str,
     summary: str,
     checklist: list[str],
+    decision_source: str = "rules",
+    materials_summary: str = "",
 ) -> None:
     conn.execute(
         """
         INSERT OR REPLACE INTO renewal_tasks
             (task_id, contract_id, alert_id, ai_suggestion, suggestion_reason,
-             performance_summary, materials_checklist, generated_at, status)
-        VALUES (?, ?, ?, ?, ?, ?, ?, ?, 'pending')
+             performance_summary, materials_checklist, generated_at, status,
+             decision_source, materials_summary)
+        VALUES (?, ?, ?, ?, ?, ?, ?, ?, 'pending', ?, ?)
         """,
         (
             task_id,
@@ -143,6 +146,8 @@ def insert_task(
             summary,
             json.dumps(checklist, ensure_ascii=False),
             datetime.now().isoformat(timespec="seconds"),
+            decision_source,
+            materials_summary,
         ),
     )
 
@@ -160,6 +165,22 @@ def get_task(conn: sqlite3.Connection, task_id: str) -> sqlite3.Row | None:
     ).fetchone()
 
 
+def get_task_full(conn: sqlite3.Connection, task_id: str) -> sqlite3.Row | None:
+    """v1.1.1：拉取 task + 关联合同 + alert（含 days_to_expire/level）。"""
+    return conn.execute(
+        """
+        SELECT t.*, c.teacher_name, c.unit_name, c.position, c.contract_type,
+               c.contract_start, c.contract_end, c.employee_id, c.unit_id,
+               a.days_to_expire AS alert_days, a.alert_level AS alert_level
+        FROM renewal_tasks t
+        JOIN contracts c ON c.contract_id = t.contract_id
+        LEFT JOIN alerts a ON a.alert_id = t.alert_id
+        WHERE t.task_id = ?
+        """,
+        (task_id,),
+    ).fetchone()
+
+
 # ---------- dispatch_log ----------
 
 def insert_dispatch_log(
@@ -193,3 +214,86 @@ def list_dispatch_logs(conn: sqlite3.Connection) -> list[sqlite3.Row]:
     return conn.execute(
         "SELECT * FROM dispatch_log ORDER BY dispatched_at DESC"
     ).fetchall()
+
+
+# ---------- materials_content（v1.1.1 新增） ----------
+
+def list_materials_for_contract(
+    conn: sqlite3.Connection, contract_id: str
+) -> list[sqlite3.Row]:
+    """列出某合同的全部材料文件。"""
+    return conn.execute(
+        "SELECT * FROM materials_content WHERE contract_id = ? ORDER BY parsed_at ASC",
+        (contract_id,),
+    ).fetchall()
+
+
+def insert_material(
+    conn: sqlite3.Connection,
+    contract_id: str,
+    file_name: str,
+    file_path: str,
+    extracted_text: str,
+    parse_method: str,
+) -> None:
+    conn.execute(
+        """
+        INSERT OR REPLACE INTO materials_content
+            (contract_id, file_name, file_path, extracted_text,
+             parse_method, parsed_at)
+        VALUES (?, ?, ?, ?, ?, ?)
+        """,
+        (
+            contract_id,
+            file_name,
+            file_path,
+            extracted_text,
+            parse_method,
+            datetime.now().isoformat(timespec="seconds"),
+        ),
+    )
+
+
+def get_material(
+    conn: sqlite3.Connection, contract_id: str, file_name: str
+) -> sqlite3.Row | None:
+    return conn.execute(
+        "SELECT * FROM materials_content WHERE contract_id = ? AND file_name = ?",
+        (contract_id, file_name),
+    ).fetchone()
+
+
+def list_pending_materials(conn: sqlite3.Connection) -> list[sqlite3.Row]:
+    """列出待解析的材料（extracted_text IS NULL）。"""
+    return conn.execute(
+        "SELECT * FROM materials_content "
+        "WHERE extracted_text IS NULL OR parse_method = 'pending' "
+        "ORDER BY parsed_at ASC"
+    ).fetchall()
+
+
+def summarize_materials(
+    conn: sqlite3.Connection, contract_id: str
+) -> tuple[str, list[str]]:
+    """汇总某合同所有已解析材料的文本 + 文件名列表。
+
+    返回 (summary_text, file_names)。
+    """
+    rows = list_materials_for_contract(conn, contract_id)
+    if not rows:
+        return ("", [])
+
+    file_names: list[str] = []
+    parts: list[str] = []
+    for r in rows:
+        text = r["extracted_text"] or ""
+        if not text:
+            continue
+        file_names.append(r["file_name"])
+        parts.append(f"### {r['file_name']}\n{text}")
+
+    summary = "\n\n".join(parts)
+    # 截断 16k 字避免 LLM context 爆栈
+    if len(summary) > 16000:
+        summary = summary[:16000] + "\n...(材料过长已截断)"
+    return (summary, file_names)