From 84d842bfcc7850cb6048bbb516a7309f64e70242 Mon Sep 17 00:00:00 2001
From: adamhenson <adamhenson1979@gmail.com>
Date: Fri, 29 May 2026 14:19:01 -0400
Subject: [PATCH 1/3] fix(log-watcher): remove branch filter; fold high-cost
 alert into primary report

---
 workflows/log-watcher.md | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/workflows/log-watcher.md b/workflows/log-watcher.md
index 0936fdc..0eecf64 100644
--- a/workflows/log-watcher.md
+++ b/workflows/log-watcher.md
@@ -10,8 +10,6 @@ on:
     workflows: ["agent-implement", "agent-pr-fix"]  # Edit to match your agent workflow names
     types:
       - completed
-    branches:
-      - main
 
 permissions: read-all
 
@@ -194,14 +192,18 @@ run is degraded or failed, the most likely cause.
 **If no PR was found**: create an issue using `create_issue` with title:
 `[log-watcher] #${{ github.event.workflow_run.run_number }}: $HEALTH`
 
-### Step 8: Critical anomaly alert (optional)
+### Step 8: High-cost failure callout (optional)
 
-If health is ❌ **Failed** AND total tokens exceed 50 000 (high-cost failure), create a
-second issue using `create_issue` with title:
-`[log-watcher] High-cost failure: run #${{ github.event.workflow_run.run_number }}`
+If health is ❌ **Failed** AND total tokens exceed 50 000, add the following callout
+**inside** the report you are already posting (comment or issue from Step 7) - do not
+create a separate issue:
 
-Include the full diagnosis and a direct link to the run. Adjust the 50 000-token threshold
-in the workflow to match your budget.
+```markdown
+> ⚠️ **High-cost failure** - this run consumed $TOTAL_TOKENS tokens. Review the token
+> breakdown above. Adjust the 50 000-token threshold in the workflow to match your budget.
+```
+
+This keeps one report per run as required by the guidelines below.
 
 ## Guidelines
 

From 03ccd95b31c4dc075e2c1b57594d800b1e9b4dc0 Mon Sep 17 00:00:00 2001
From: adamhenson <adamhenson1979@gmail.com>
Date: Fri, 29 May 2026 14:29:21 -0400
Subject: [PATCH 2/3] docs(log-watcher): document optional branch filter with
 trade-off comment

---
 workflows/log-watcher.md | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/workflows/log-watcher.md b/workflows/log-watcher.md
index 0eecf64..d3c8d83 100644
--- a/workflows/log-watcher.md
+++ b/workflows/log-watcher.md
@@ -10,6 +10,9 @@ on:
     workflows: ["agent-implement", "agent-pr-fix"]  # Edit to match your agent workflow names
     types:
       - completed
+    # branches: [main]  # Uncomment to restrict to specific branches.
+    #                   # Omit (default) to watch all branches, including PR branches,
+    #                   # which is the common gh-aw case.
 
 permissions: read-all
 

From aae19cea77cd4c70fcbcf601826118860f72e5f5 Mon Sep 17 00:00:00 2001
From: adamhenson <adamhenson1979@gmail.com>
Date: Fri, 29 May 2026 14:35:58 -0400
Subject: [PATCH 3/3] docs(log-watcher): define $TOTAL_TOKENS capture in Step 4
 for use in Step 8

---
 workflows/log-watcher.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/workflows/log-watcher.md b/workflows/log-watcher.md
index d3c8d83..5a99345 100644
--- a/workflows/log-watcher.md
+++ b/workflows/log-watcher.md
@@ -133,6 +133,10 @@ Calculate the following metrics across all lines:
 
 Flagged metrics are anomalies - include them in the diagnosis.
 
+Capture the total token count as `$TOTAL_TOKENS` (sum of all `input_tokens`, `output_tokens`,
+`cache_read_input_tokens`, and `cache_creation_input_tokens` across all lines) for use in
+Step 8.
+
 ### Step 5: Determine run health
 
 Assign one of three health levels: