epimetheus: fix double research message + add decisions/ to KB retrieval

1. handle_research gets silent=True param. RESEARCH: tag triggers use
   silent mode — archives tweets but posts no follow-up message.
   Prevents "Queued N tweets" after Opus already responded.

2. KB retrieval now searches decisions/ directory alongside domains/,
   core/, foundations/. Decision records (Robin Hanson proposal, etc.)
   are now findable by the bot.

Pentagon-Agent: Epimetheus <3D35839A-7722-4740-B93D-51157F7D5E70>

telegram/bot.py

@@ -276,22 +276,28 @@ def _format_conversation_history(chat_id: int, user_id: int) -> str:
 RESEARCH_PATTERN = re.compile(r'/research(?:@\w+)?\s+(.+)', re.IGNORECASE)
 
 
-async def handle_research(msg, query: str, user):
-    """Handle a research request — search X and archive results as sources."""
+async def handle_research(msg, query: str, user, silent: bool = False):
+    """Handle a research request — search X and archive results as sources.
+
+    If silent=True, archive only — no messages posted. Used when triggered
+    by RESEARCH: tag after Opus already responded.
+    """
     username = user.username if user else "unknown"
 
-    if not check_research_rate_limit(user.id if user else 0):
+    if not silent and not check_research_rate_limit(user.id if user else 0):
         remaining = get_research_remaining(user.id if user else 0)
         await msg.reply_text(f"Research limit reached (3/day). Resets at midnight UTC. {remaining} remaining.")
         return
 
-    await msg.chat.send_action("typing")
+    if not silent:
+        await msg.chat.send_action("typing")
 
     logger.info("Research: searching X for '%s'", query)
     tweets = await search_tweets(query, max_results=15, min_engagement=0)
     logger.info("Research: got %d tweets for '%s'", len(tweets), query)
     if not tweets:
-        await msg.reply_text(f"No recent tweets found for '{query}'.")
+        if not silent:
+            await msg.reply_text(f"No recent tweets found for '{query}'.")
         return
 
     # Archive all tweets as ONE source file per research query
@@ -343,18 +349,17 @@ Submitted by @{username} via Telegram /research command.
     except Exception as e:
         logger.warning("Research archive failed: %s", e)
 
-    record_research_usage(user.id if user else 0)
-    remaining = get_research_remaining(user.id if user else 0)
-
-    # Summary of what was found
-    top_authors = list(set(t["author"] for t in tweets[:5]))
-    await msg.reply_text(
-        f"Queued {archived} tweets about '{query}' for extraction. "
-        f"Top voices: @{', @'.join(top_authors[:3])}. "
-        f"Results will appear in the KB within ~30 minutes. "
-        f"({remaining} research requests remaining today.)"
-    )
-    logger.info("Research: @%s queried '%s', archived %d tweets", username, query, archived)
+    if not silent:
+        record_research_usage(user.id if user else 0)
+        remaining = get_research_remaining(user.id if user else 0)
+        top_authors = list(set(t["author"] for t in tweets[:5]))
+        await msg.reply_text(
+            f"Queued {archived} tweets about '{query}' for extraction. "
+            f"Top voices: @{', @'.join(top_authors[:3])}. "
+            f"Results will appear in the KB within ~30 minutes. "
+            f"({remaining} research requests remaining today.)"
+        )
+    logger.info("Research: @%s queried '%s', archived %d tweets (silent=%s)", username, query, archived, silent)
 
 
 # ─── Message Handlers ───────────────────────────────────────────────────
@@ -644,7 +649,7 @@ IMPORTANT: Two special tags you can append at the end of your response (after yo
         display_response = re.sub(r'\nRESEARCH:\s+.+$', '', display_response, flags=re.MULTILINE).rstrip()
         if not research_context:  # Only fire if Haiku didn't already search
             for query in research_lines:
-                asyncio.get_event_loop().create_task(handle_research(msg, query.strip(), user))
+                asyncio.get_event_loop().create_task(handle_research(msg, query.strip(), user, silent=True))
                 logger.info("Auto-research triggered: %s", query[:80])
 
     # Post response (without LEARNING lines)

telegram/kb_retrieval.py

@@ -185,6 +185,7 @@ class KBIndex:
             self.repo_dir / "domains",
             self.repo_dir / "core",
             self.repo_dir / "foundations",
+            self.repo_dir / "decisions",
         ]
         for claim_dir in claim_dirs:
             if not claim_dir.exists():