rapid7-insightidr 1.14.8: user_authentications counts the user's own logons

The keyword search matches the account name anywhere in a line, so it brought
other accounts' logons: on a real run the only "new" event reported was
another user's failed password. Only events whose account/user field is the
user (DOMAIN\\user and user@domain normalized) are summarized now, with
keyword_matches and user_matches both reported. A first authentication is
recognized as Rapid7 flags it, a boolean new_authentication field, as well as
the word. The event limit cannot be set below 10000 (the agent passed 1000).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015RhL2gy6Sx6Y3FEsyBE3mZ

integrations/rapid7-insightidr/scripts/user_authentications.py

@@ -255,6 +255,42 @@ def _resolve_name(name):
 
 
 NEW_MARKERS = ("new_authentication",)
+_TRUE = ("true", "1", "yes")
+
+
+def _norm_account(v):
+    """jdoe, DOMAIN\\jdoe, jdoe@domain.tld -> jdoe."""
+    v = str(v or "").strip().lower()
+    if "\\" in v:
+        v = v.rsplit("\\", 1)[1]
+    if "@" in v:
+        v = v.split("@", 1)[0]
+    return v
+
+
+def _is_users(fields, user):
+    """The event is this user's: an account or user field holds the account.
+
+    The keyword search matches the text anywhere in the line, so it also brings
+    other people's logons — on a real run the only "new" event it found was
+    another account's. Only the account fields decide."""
+    want = _norm_account(user)
+    for k, v in fields.items():
+        kl = k.lower()
+        if ("account" in kl or "user" in kl) and "sid" not in kl and _norm_account(v) == want:
+            return True
+    return False
+
+
+def _is_new(fields):
+    """Rapid7 marks a first-time authentication either with a boolean field
+    (new_authentication: true) or with that word as a value."""
+    for k, v in fields.items():
+        if "new_authentication" in k.lower() and str(v).strip().lower() in _TRUE:
+            return True
+        if str(v).strip().lower() in NEW_MARKERS:
+            return True
+    return False
 
 
 def run():
@@ -278,24 +314,29 @@ def run():
     name = str(inp.get("log_set_name") or "Asset Authentication").strip()
     log_set_id = _resolve_name(name)
     try:
-        limit = int(inp.get("limit") or 2000)
+        limit = max(int(inp.get("limit") or 10000), 10000)
     except (TypeError, ValueError):
-        limit = 2000
+        limit = 10000
     query = 'where("%s")' % user
     first = request("GET", "log_search/query/logsets/" + urllib.parse.quote(log_set_id, safe=""),
                     params={"query": query, "from": start_time, "to": end_time})
-    events = _identify(_collect(first, limit), "", None)
-    new = [e for e in events
-           if any(str(v).lower() in NEW_MARKERS for v in _fields(e).values())]
+    matched = _identify(_collect(first, limit), "", None)
+    events = [e for e in matched if _is_users(_fields(e), user)]
+    new = [e for e in events if _is_new(_fields(e))]
     print(json.dumps({
         "user": user,
         "log_set": name,
         "query": query,
+        "keyword_matches": len(matched),
+        "user_matches": len(events),
+        "truncated": len(matched) >= limit,
         "window": "%s to %s" % (_iso(start_time), _iso(end_time)),
         "all": _summarize(events, limit),
         "new_authentications": _summarize(new, limit),
-        "note": ("new_authentications holds only the events Rapid7 marked new_authentication: their "
-                 "destination assets are the new assets a behaviour alert counts."),
+        "note": ("all and new_authentications hold only this user's events (an account field equals "
+                 "the user): keyword_matches counts every line the text appeared in, other accounts "
+                 "included. new_authentications holds the events Rapid7 marked as a first "
+                 "authentication: their destination assets are the new assets a behaviour alert counts."),
     }))