feat: add _sanitize_mentions helper and apply to comment/reply truncation

Introduce a new `_sanitize_mentions` method in `BotHelpersMixin` that deduplicates and removes self-mentions from text, using a compiled regex for `@handle` patterns. Apply this sanitizer before the 2000-character truncation in both `engage.py` comment posting and `social.py` reply posting to prevent duplicate or self-referential mentions from being cut off mid-handle. Additionally, fix profile URL parsing in `social.py` to strip trailing path segments, and refine `LLMClient.clean` to handle asterisks and underscores more precisely without breaking adjacent alphanumeric characters.

devplacepy/services/bot/engage.py

@@ -305,7 +305,7 @@ class BotEngageMixin:
             else:
                 comment = f"@{mention_target} {comment}"
 
-        comment = comment[:2000]
+        comment = self._sanitize_mentions(comment)[:2000]
         mention_log = f" @{mention_target}" if mention_target else ""
 
         textarea_sels = [
@@ -437,7 +437,7 @@ class BotEngageMixin:
                 return False
             if mentioner and f"@{mentioner.lower()}" not in reply.lower():
                 reply = f"@{mentioner} {reply}"
-            reply = reply[:2000]
+            reply = self._sanitize_mentions(reply)[:2000]
             textarea_sels = [
                 ".reply-form textarea[name='content']",
                 ".comment-form textarea[name='content']",

devplacepy/services/bot/helpers.py

@@ -11,6 +11,8 @@ from devplacepy.services.bot.config import persona_article_score, pick_category
 
 logger = logging.getLogger(__name__)
 
+_MENTION_RE = re.compile(r"@([A-Za-z0-9_-]+)")
+
 
 class BotHelpersMixin:
     def _identity(self) -> str:
@@ -31,6 +33,21 @@ class BotHelpersMixin:
         m = re.search(r"/(?:posts|gists|news|projects)/([^/?#]+)", url)
         return m.group(1) if m else ""
 
+    def _sanitize_mentions(self, text: str) -> str:
+        own = (self.state.username or "").lower()
+        seen: set[str] = set()
+
+        def keep(match: "re.Match[str]") -> str:
+            handle = match.group(1)
+            lowered = handle.lower()
+            if lowered == own or lowered in seen:
+                return ""
+            seen.add(lowered)
+            return match.group(0)
+
+        cleaned = _MENTION_RE.sub(keep, text)
+        return re.sub(r"\s{2,}", " ", cleaned).strip()
+
     def _sync_cost(self) -> None:
         self.state.total_cost = self.llm.total_cost
         self.state.total_calls = self.llm.total_calls

devplacepy/services/bot/llm.py

@@ -115,7 +115,10 @@ class LLMClient:
     @staticmethod
     def clean(text: str, preserve_md: bool = False) -> str:
         if not preserve_md:
-            text = re.sub(r"\*\*|__|\*|_", "", text)
+            text = re.sub(r"\*+", "", text)
+            text = re.sub(r"(?<![A-Za-z0-9])__(?=\S)(.*?)(?<=\S)__(?![A-Za-z0-9])", r"\1", text)
+            text = re.sub(r"(?<![A-Za-z0-9])_(?=\S)(.*?)(?<=\S)_(?![A-Za-z0-9])", r"\1", text)
+            text = re.sub(r"(?<![A-Za-z0-9])_+(?![A-Za-z0-9])", "", text)
         text = text.replace("—", "-").replace("–", "-")
         text = text.replace("‘", "'").replace("’", "'")
         text = text.replace("“", '"').replace("”", '"')

devplacepy/services/bot/social.py

@@ -28,7 +28,7 @@ class BotSocialMixin:
                 try:
                     await links.nth(i).click(timeout=5000)
                     await b._idle(1.0, 2.0)
-                    uname = h.split("/profile/")[-1].split("?")[0]
+                    uname = h.split("/profile/")[-1].split("?")[0].split("/")[0]
                     if uname not in self.state.known_users:
                         self.state.known_users.append(uname)
                     self.state.profiles_viewed += 1
@@ -446,6 +446,9 @@ class BotSocialMixin:
                         )
             except Exception:
                 mentioner = ""
+            if mentioner and mentioner.lower() == self.state.username.lower():
+                self._log("Mention target resolves to self, skipping")
+                return False
             try:
                 await comment_loc.scroll_into_view_if_needed(timeout=2000)
                 await b._idle(0.3, 0.9)
@@ -485,7 +488,7 @@ class BotSocialMixin:
             return False
         if mentioner and f"@{mentioner.lower()}" not in reply.lower():
             reply = f"@{mentioner} {reply}"
-        reply = reply[:2000]
+        reply = self._sanitize_mentions(reply)[:2000]
         textarea_sels = [
             ".comment .reply-form textarea[name='content']",
             ".reply-form textarea[name='content']",