fix: clear pending tool-call state across provider modes (#32120) (thanks @jnMetaCode)

CHANGELOG.md

@@ -62,6 +62,7 @@ Docs: https://docs.openclaw.ai
 - Config/backups hardening: enforce owner-only (`0600`) permissions on rotated config backups and clean orphan `.bak.*` files outside the managed backup ring, reducing credential leakage risk from stale or permissive backup artifacts. (#31718) Thanks @YUJIE2002.
 - WhatsApp/inbound self-message context: propagate inbound `fromMe` through the web inbox pipeline and annotate direct self messages as `(self)` in envelopes so agents can distinguish owner-authored turns from contact turns. (#32167) Thanks @scoootscooob.
 - Exec approvals/allowlist matching: escape regex metacharacters in path-pattern literals (while preserving glob wildcards), preventing crashes on allowlisted executables like `/usr/bin/g++` and correctly matching mixed wildcard/literal token paths. (#32162) Thanks @stakeswky.
+- Agents/tool-result guard: always clear pending tool-call state on interruptions even when synthetic tool results are disabled, preventing orphaned tool-use transcripts that cause follow-up provider request failures. (#32120) Thanks @jnMetaCode.
 - Webchat/stream finalization: persist streamed assistant text when final events omit `message`, while keeping final payload precedence and skipping empty stream buffers to prevent disappearing replies after tool turns. (#31920) Thanks @Sid-Qin.
 - Cron/store migration: normalize legacy cron jobs with string `schedule` and top-level `command`/`timeout` fields into canonical schedule/payload/session-target shape on load, preventing schedule-error loops on old persisted stores. (#31926) Thanks @bmendonca3.
 - Gateway/Heartbeat model reload: treat `models.*` and `agents.defaults.model` config updates as heartbeat hot-reload triggers so heartbeat picks up model changes without a full gateway restart. (#32046) Thanks @stakeswky.

src/agents/session-tool-result-guard.test.ts

@@ -85,6 +85,25 @@ describe("installSessionToolResultGuard", () => {
     expectPersistedRoles(sm, ["assistant", "toolResult"]);
   });
 
+  it("clears pending on user interruption when synthetic tool results are disabled", () => {
+    const sm = SessionManager.inMemory();
+    const guard = installSessionToolResultGuard(sm, {
+      allowSyntheticToolResults: false,
+    });
+
+    sm.appendMessage(toolCallMessage);
+    sm.appendMessage(
+      asAppendMessage({
+        role: "user",
+        content: "interrupt",
+        timestamp: Date.now(),
+      }),
+    );
+
+    expectPersistedRoles(sm, ["assistant", "user"]);
+    expect(guard.getPendingIds()).toEqual([]);
+  });
+
   it("does not add synthetic toolResult when a matching one exists", () => {
     const sm = SessionManager.inMemory();
     installSessionToolResultGuard(sm);
@@ -271,6 +290,54 @@ describe("installSessionToolResultGuard", () => {
     expectPersistedRoles(sm, ["assistant", "toolResult"]);
   });
 
+  it("clears pending when a sanitized assistant message is dropped and synthetic results are disabled", () => {
+    const sm = SessionManager.inMemory();
+    const guard = installSessionToolResultGuard(sm, {
+      allowSyntheticToolResults: false,
+      allowedToolNames: ["read"],
+    });
+
+    sm.appendMessage(
+      asAppendMessage({
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_1", name: "read", arguments: {} }],
+      }),
+    );
+
+    sm.appendMessage(
+      asAppendMessage({
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_2", name: "write", arguments: {} }],
+      }),
+    );
+
+    expectPersistedRoles(sm, ["assistant"]);
+    expect(guard.getPendingIds()).toEqual([]);
+  });
+
+  it("drops older pending ids before new tool calls when synthetic results are disabled", () => {
+    const sm = SessionManager.inMemory();
+    const guard = installSessionToolResultGuard(sm, {
+      allowSyntheticToolResults: false,
+    });
+
+    sm.appendMessage(
+      asAppendMessage({
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_1", name: "read", arguments: {} }],
+      }),
+    );
+    sm.appendMessage(
+      asAppendMessage({
+        role: "assistant",
+        content: [{ type: "toolCall", id: "call_2", name: "read", arguments: {} }],
+      }),
+    );
+
+    expectPersistedRoles(sm, ["assistant", "assistant"]);
+    expect(guard.getPendingIds()).toEqual(["call_2"]);
+  });
+
   it("caps oversized tool result text during persistence", () => {
     const sm = SessionManager.inMemory();
     installSessionToolResultGuard(sm);