Add Discord attachment reading and web search capabilities

- Discord channel now downloads and extracts text from attachments (text files, PDFs)
- Added WebSearchTool using DuckDuckGo for researcher and coder agents
- Improved WebFetchTool with User-Agent header and HTML-to-text stripping
- Added pypdf and duckduckgo-search dependencies

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

xtrm_agent/channels/discord.py

@@ -3,14 +3,29 @@
 from __future__ import annotations
 
 import asyncio
+import io
 import os
 
 import discord
+import httpx
 from loguru import logger
 
-from xtrm_agent.bus import InboundMessage, MessageBus, OutboundMessage
+from xtrm_agent.bus import Attachment, InboundMessage, MessageBus, OutboundMessage
 from xtrm_agent.channels.base import BaseChannel
 
+# Extensions treated as plain text (decoded as UTF-8)
+_TEXT_EXTENSIONS = frozenset({
+    ".txt", ".py", ".md", ".json", ".yaml", ".yml", ".csv", ".log",
+    ".js", ".ts", ".html", ".css", ".xml", ".toml", ".ini", ".sh",
+    ".sql", ".rs", ".go", ".java", ".c", ".cpp", ".h", ".rb", ".php",
+    ".swift", ".kt", ".r", ".cfg", ".env", ".conf", ".dockerfile",
+    ".makefile", ".bat", ".ps1", ".lua", ".zig", ".hs",
+})
+
+_IMAGE_EXTENSIONS = frozenset({".png", ".jpg", ".jpeg", ".gif", ".webp", ".bmp", ".svg"})
+
+_MAX_ATTACHMENT_SIZE = 1_024_000  # 1 MB
+
 
 class DiscordChannel(BaseChannel):
     """Discord bot channel."""
@@ -54,12 +69,15 @@ class DiscordChannel(BaseChannel):
             if self.client.user:
                 content = content.replace(f"<@{self.client.user.id}>", "").strip()
 
+            attachments = await self._extract_attachments(message.attachments)
+
             msg = InboundMessage(
                 channel="discord",
                 sender_id=str(message.author.id),
                 chat_id=str(message.channel.id),
                 content=content,
                 metadata={"guild_id": str(message.guild.id) if message.guild else ""},
+                attachments=attachments,
             )
             await self.bus.publish_inbound(msg)
 
@@ -71,6 +89,80 @@ class DiscordChannel(BaseChannel):
             except asyncio.TimeoutError:
                 await message.channel.send("Sorry, I timed out processing your request.")
 
+    async def _extract_attachments(
+        self, discord_attachments: list[discord.Attachment]
+    ) -> list[Attachment]:
+        """Download Discord attachments and extract text content."""
+        results: list[Attachment] = []
+        for att in discord_attachments:
+            name = att.filename.lower()
+            ext = "." + name.rsplit(".", 1)[-1] if "." in name else ""
+
+            if att.size > _MAX_ATTACHMENT_SIZE:
+                results.append(Attachment(
+                    filename=att.filename,
+                    content=f"(file skipped — {att.size / 1_048_576:.1f} MB exceeds 1 MB limit)",
+                ))
+                continue
+
+            if ext in _IMAGE_EXTENSIONS:
+                results.append(Attachment(
+                    filename=att.filename,
+                    content="(image attached — cannot read image content)",
+                ))
+                continue
+
+            try:
+                async with httpx.AsyncClient(timeout=30) as client:
+                    resp = await client.get(att.url)
+                    resp.raise_for_status()
+                    raw = resp.content
+            except Exception as e:
+                logger.warning(f"Failed to download attachment {att.filename}: {e}")
+                results.append(Attachment(
+                    filename=att.filename,
+                    content=f"(failed to download: {e})",
+                ))
+                continue
+
+            if ext == ".pdf":
+                try:
+                    from pypdf import PdfReader
+
+                    reader = PdfReader(io.BytesIO(raw))
+                    text = "\n".join(
+                        page.extract_text() or "" for page in reader.pages
+                    ).strip()
+                    if text:
+                        results.append(Attachment(filename=att.filename, content=text))
+                    else:
+                        results.append(Attachment(
+                            filename=att.filename,
+                            content="(PDF has no extractable text)",
+                        ))
+                except Exception as e:
+                    logger.warning(f"Failed to extract PDF text from {att.filename}: {e}")
+                    results.append(Attachment(
+                        filename=att.filename,
+                        content=f"(failed to read PDF: {e})",
+                    ))
+            elif ext in _TEXT_EXTENSIONS or (att.content_type and att.content_type.startswith("text/")):
+                try:
+                    text = raw.decode("utf-8", errors="replace")
+                    results.append(Attachment(filename=att.filename, content=text))
+                except Exception as e:
+                    results.append(Attachment(
+                        filename=att.filename,
+                        content=f"(failed to decode text: {e})",
+                    ))
+            else:
+                results.append(Attachment(
+                    filename=att.filename,
+                    content=f"(unsupported file type: {ext or 'unknown'})",
+                ))
+
+        return results
+
     async def _send_chunked(
         self, channel: discord.abc.Messageable, content: str
     ) -> None: