Add digest command to email processor

Read-only summary of recent decisions, grouped by action with
[auto]/[user] markers. Supports --recent N for multi-day lookback.

TOOLS.md

@@ -89,6 +89,8 @@ agent-browser close
 ./email-processor.sh review accept           # 接受所有建议
 ./email-processor.sh review 1 delete         # 处理第1封（删除）
 ./email-processor.sh stats                   # 查看统计
+./email-processor.sh digest                  # 今日处理摘要
+./email-processor.sh digest --recent 3       # 最近3天摘要
 ```
 
 **置信度机制**:

scripts/email_processor/README.md

@@ -60,6 +60,7 @@ The system separates **classification** (what the LLM does) from **confidence**
 - **Repeat senders** with consistent tag signatures reach 85%+ confidence and get auto-acted during `scan`. They never touch the pending queue.
 - **New or ambiguous senders** start at 50% and get queued.
 - **You occasionally run `review list`** to handle stragglers — each decision further builds history.
+- **`digest` gives a quick glance** at what was processed recently — subject lines grouped by action, with `[auto]`/`[user]` markers.
 - **`stats` shows your automation rate** climbing over time.
 
 ### Confidence Computation
@@ -96,6 +97,10 @@ chmod +x email-processor.sh
 ./email-processor.sh review all delete             # delete all pending
 ./email-processor.sh review accept                 # accept all suggestions
 
+# --- Digest ---
+./email-processor.sh digest                       # today's processed emails
+./email-processor.sh digest --recent 3            # last 3 days
+
 # --- Other ---
 ./email-processor.sh stats                        # show decision history
 ```
@@ -244,13 +249,16 @@ ollama list  # should show kamekichi128/qwen3-4b-instruct-2507:latest
 
 # 7. Check that the decision was recorded
 ./email-processor.sh stats
+
+# 8. Quick glance at what was processed today
+./email-processor.sh digest
 ```
 
 ## File Structure
 
 ```
 email_processor/
-  main.py              # Entry point — scan/review/stats subcommands
+  main.py              # Entry point — scan/review/stats/digest subcommands
   classifier.py        # LLM prompt builder + response parser, tag taxonomy
   decision_store.py    # Decision history, confidence computation, few-shot retrieval
   config.json          # Ollama + automation settings

scripts/email_processor/decision_store.py

@@ -14,7 +14,7 @@ auto decision.
 
 import json
 import re
-from datetime import datetime
+from datetime import datetime, timedelta
 from pathlib import Path
 from collections import Counter
 
@@ -200,6 +200,33 @@ def get_known_labels():
     return labels
 
 
+def get_recent_decisions(days=1):
+    """Return recent decisions grouped by action.
+
+    Args:
+        days: number of days to look back (default 1 = today).
+
+    Returns:
+        dict of action -> list of entries, e.g. {"delete": [...], "archive": [...]}.
+        Returns empty dict if no decisions found in the period.
+    """
+    history = _load_history()
+    if not history:
+        return {}
+
+    cutoff = datetime.now() - timedelta(days=days)
+    grouped = {}
+    for entry in history:
+        try:
+            ts = datetime.fromisoformat(entry["timestamp"])
+        except (KeyError, ValueError):
+            continue
+        if ts >= cutoff:
+            action = entry.get("action", "unknown")
+            grouped.setdefault(action, []).append(entry)
+    return grouped
+
+
 def get_all_stats():
     """Compute aggregate statistics across the full decision history.
 

scripts/email_processor/email-processor.sh

@@ -11,6 +11,8 @@
 #   ./email-processor.sh review all delete              # act on all pending
 #   ./email-processor.sh review accept                  # accept all suggestions
 #   ./email-processor.sh stats                          # show history stats
+#   ./email-processor.sh digest                         # today's processed emails
+#   ./email-processor.sh digest --recent 3              # last 3 days
 #
 # Requires: uv, himalaya, Ollama running with model.
 

scripts/email_processor/main.py

@@ -19,6 +19,8 @@ Subcommands:
     python main.py review all <action>               # act on all pending
     python main.py review accept                     # accept all suggestions
     python main.py stats                             # show decision history
+    python main.py digest                            # today's processed emails
+    python main.py digest --recent 3                 # last 3 days
 
 Action mapping (what each classification does to the email):
     delete    -> himalaya message delete <id>  (moves to Trash)
@@ -633,6 +635,56 @@ def cmd_stats():
         print(f"\nKnown labels: {', '.join(sorted(labels))}")
 
 
+# ---------------------------------------------------------------------------
+# Subcommand: digest
+# ---------------------------------------------------------------------------
+
+def cmd_digest(days=1):
+    """Print a compact summary of recently processed emails, grouped by action.
+
+    Shows both auto and user decisions with a marker to distinguish them.
+    """
+    grouped = decision_store.get_recent_decisions(days)
+
+    if not grouped:
+        period = "today" if days == 1 else f"last {days} days"
+        print(f"No processed emails in this period ({period}).")
+        return
+
+    period_label = "today" if days == 1 else f"last {days} days"
+    print(f"Email digest ({period_label})")
+    print("=" * 40)
+
+    # Map action names to display labels
+    action_labels = {
+        "delete": "Deleted",
+        "archive": "Archived",
+        "keep": "Kept",
+        "mark_read": "Marked read",
+    }
+
+    total = 0
+    auto_count = 0
+    user_count = 0
+
+    for action, entries in sorted(grouped.items()):
+        label = action_labels.get(action, action.replace("label:", "Labeled ").title())
+        print(f"\n{label} ({len(entries)}):")
+        for entry in entries:
+            source = entry.get("source", "?")
+            sender = entry.get("sender", "unknown")
+            subject = entry.get("subject", "(no subject)")
+            print(f"  [{source}] {sender}")
+            print(f"         {subject}")
+            total += 1
+            if source == "auto":
+                auto_count += 1
+            else:
+                user_count += 1
+
+    print(f"\nTotal: {total} emails processed ({auto_count} auto, {user_count} user)")
+
+
 # ---------------------------------------------------------------------------
 # Entry point & argument parsing
 #
@@ -695,7 +747,10 @@ if __name__ == "__main__":
     elif subcommand == "stats":
         cmd_stats()
 
+    elif subcommand == "digest":
+        cmd_digest(days=recent if recent else 1)
+
     else:
         print(f"Unknown subcommand: {subcommand}")
-        print("Usage: python main.py [scan|review|stats] [--recent N] [--dry-run]")
+        print("Usage: python main.py [scan|review|stats|digest] [--recent N] [--dry-run]")
         sys.exit(1)