diff --git a/.gitignore b/.gitignore
index 87880db7..8b416d85 100644
--- a/.gitignore
+++ b/.gitignore
@@ -203,13 +203,6 @@ docs/launch-post.md
 docs/public-launch-narrative.md
 docs/RELEASE-*-DRAFT.md
 
-# Orphaned dirs from S105 splits — cloud/ moved to private
-# Gradata/gradata-cloud in #76, sdk/ superseded by the flattened repo
-# layout in #65. Left on disk as untracked copies that keep showing
-# up in `git status` and sometimes get accidentally re-added.
-/cloud/
-/sdk/
-
 # Railway config for the cloud API now lives in the private
 # gradata-cloud repo at cloud/railway.toml — the root-level copy
 # here is a stale leftover from before the split.
diff --git a/src/gradata/brain.py b/src/gradata/brain.py
index baf3e230..86e231d6 100644
--- a/src/gradata/brain.py
+++ b/src/gradata/brain.py
@@ -916,18 +916,6 @@ def apply_brain_rules(
 
         result = format_rules_for_prompt(applied)
         self._rule_cache.put(cache_key, result)
-
-        # Fires the in-memory bus so SessionHistory (integrations/session_history.py)
-        # can track per-session rule effectiveness — it subscribes to rules.injected
-        # but the emitter was never wired, leaving compute_effectiveness() dormant.
-        # Cache hits above skip this intentionally: SessionHistory uses a set so
-        # replays are idempotent within a session, and the cache is per-scope
-        # so repeated injects in the same scope yield the same rule set.
-        if applied:
-            self.bus.emit("rules.injected", {
-                "rules": [{"id": a.rule_id} for a in applied],
-                "task": task,
-            })
         return result
 
     def scoped_rules(