From f418afa2756e887d1428b52daa98ff3ed0520276 Mon Sep 17 00:00:00 2001
From: Lukasz Raczylo <lukasz@raczylo.com>
Date: Fri, 25 Sep 2026 03:16:01 +0100
Subject: [PATCH] feat(observe): record real tool failures from
 PostToolUseFailure

Claude Code sends failed tool calls to the PostToolUseFailure event.
ADAM registered only PostToolUse, so it never saw a real failure, and
every tool_error_loop came from ERROR_RE matching successful output.

Handle PostToolUseFailure in adam-observe.mjs through the PostToolUse
detectors. Shape the error string as an is_error response. An interrupt
is an abort, so it counts as a tool event but gets no fingerprint.
Register the event in settings.json.example.

Tests 9b and 9c added: 150 passed, 0 failed.

Claude-Session: https://claude.ai/code/session_014prTNuJnKuwcAd28vUZAnN
---
 adam/tests/run-tests.sh | 19 +++++++++++++++++++
 hooks/adam-observe.mjs  | 15 ++++++++++-----
 settings.json.example   | 11 +++++++++++
 3 files changed, 40 insertions(+), 5 deletions(-)

diff --git a/adam/tests/run-tests.sh b/adam/tests/run-tests.sh
index 023a4b2..2767d88 100755
--- a/adam/tests/run-tests.sh
+++ b/adam/tests/run-tests.sh
@@ -187,6 +187,25 @@ for i in 1 2 3; do
 done
 assert_grep "$ROOT/journal.jsonl" '"type":"tool_error_loop"' "3x same error logs tool_error_loop"
 
+# --- Test 9b: PostToolUseFailure feeds the same detectors ---
+echo "Test 9b: 3x PostToolUseFailure with the same error logs tool_error_loop"
+reset_state
+for i in 1 2 3; do
+  echo '{"hook_event_name":"PostToolUseFailure","tool_name":"Bash","tool_input":{"command":"npm test"},"error":"Exit code 1\nError: Cannot find module express","is_interrupt":false,"session_id":"s9b","cwd":"/tmp/x"}' \
+    | HOOK_RUN >/dev/null 2>&1 || true
+done
+assert_grep "$ROOT/journal.jsonl" '"type":"tool_error_loop"' "3x PostToolUseFailure logs tool_error_loop"
+
+# --- Test 9c: interrupted tool calls are not errors ---
+echo "Test 9c: 8x PostToolUseFailure with is_interrupt:true counts as tool use but logs no tool_error_loop"
+reset_state
+for i in 1 2 3 4 5 6 7 8; do
+  echo '{"hook_event_name":"PostToolUseFailure","tool_name":"Bash","tool_input":{"command":"sleep 99"},"error":"Error: aborted","is_interrupt":true,"session_id":"s9c","cwd":"/tmp/x"}' \
+    | HOOK_RUN >/dev/null 2>&1 || true
+done
+assert_grep "$ROOT/journal.jsonl" '"type":"dead_end"' "interrupted calls still count as tool events"
+assert_no_grep "$ROOT/journal.jsonl" '"type":"tool_error_loop"' "interrupts do not log tool_error_loop"
+
 # --- Test 10: dead_end on long autonomous run ---
 echo "Test 10: dead_end after 8 tools without UserPromptSubmit"
 reset_state
diff --git a/hooks/adam-observe.mjs b/hooks/adam-observe.mjs
index b99bb47..bc546c9 100755
--- a/hooks/adam-observe.mjs
+++ b/hooks/adam-observe.mjs
@@ -465,12 +465,17 @@ function main() {
         state.subagent_dispatch_emitted[name] = true;
       }
     }
-  } else if (event === "PostToolUse") {
+  } else if (event === "PostToolUse" || event === "PostToolUseFailure") {
     const tool = input.tool_name || "unknown";
     const argsHash = djb2(JSON.stringify(input.tool_input || {}));
     const file = (input.tool_input && (input.tool_input.file_path || input.tool_input.path)) || null;
 
-    const toolResponse = input.tool_response;
+    // Claude Code sends failed tool calls to PostToolUseFailure with the error
+    // text in `error`. Shape it like an is_error response so every PostToolUse
+    // detector sees real failures. An interrupt is an abort, not a tool error.
+    const toolResponse = event === "PostToolUseFailure"
+      ? { is_error: input.is_interrupt !== true, content: String(input.error || "") }
+      : input.tool_response;
     const respExcerpt = (() => {
       if (!toolResponse) return null;
       const text = typeof toolResponse === "string" ? toolResponse
@@ -541,11 +546,11 @@ function main() {
       }
     }
 
-    if (input.tool_response && typeof input.tool_response === "object") {
+    if (toolResponse && typeof toolResponse === "object") {
       bumpUsage("payload:tool_response_seen");
     }
 
-    const fp = errorFingerprint(input.tool_response);
+    const fp = errorFingerprint(toolResponse);
     if (fp) {
       bumpUsage("payload:tool_response_error_seen");
       if (state.recentRecoveries.length) {
@@ -590,7 +595,7 @@ function main() {
     if (tool === "Bash") {
       const cmd = (input.tool_input && input.tool_input.command) || "";
       const isBuildCmd = BUILD_RE.test(cmd);
-      const hasError = (input.tool_response && input.tool_response.is_error === true) || fp !== null;
+      const hasError = (toolResponse && toolResponse.is_error === true) || fp !== null;
       if (isBuildCmd && hasError) {
         state.build_failure_count += 1;
         if (state.build_failure_count >= BUILD_LOOP_THRESHOLD && !state.build_loop_emitted) {
diff --git a/settings.json.example b/settings.json.example
index 7ce66d2..c5c420f 100644
--- a/settings.json.example
+++ b/settings.json.example
@@ -34,6 +34,17 @@
         ]
       }
     ],
+    "PostToolUseFailure": [
+      {
+        "matcher": "",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "node \"$HOME/.claude/hooks/adam-observe.mjs\""
+          }
+        ]
+      }
+    ],
     "SessionStart": [
       {
         "hooks": [