diff --git a/AGENTS.md b/AGENTS.md
index c422ed4b9a51..a3355921e1ff 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -236,6 +236,7 @@ Skills own workflows; root owns hard policy and routing.
 - Trusted-workflow release-branch CI: pass `target_ref` + `release_candidate_ref`; never `release_gate` (requires workflow head == target).
 - Agent PR landing to `main`: use only the repo-native `scripts/pr` wrapper: run `scripts/pr review-init <PR>`, follow its emitted checkout/guard guidance, initialize and complete review artifacts with `scripts/pr review-artifacts-init <PR>`, validate them with `scripts/pr review-validate-artifacts <PR>`, then run `OPENCLAW_TESTBOX=1 scripts/pr prepare-run <PR>` and `scripts/pr merge-run <PR>`. The Testbox flag is mandatory for agents so prepare verifies hosted CI/Testbox on the current head or reuses a patch-identical pre-rebase run green within 24 hours instead of running full gates locally. `prepare-run` fails fast; invoke only after exact-head CI is complete and green. For owner-approved reviewed fork code without hosted Testbox, use `OPENCLAW_PR_GATES_REMOTE=testbox` instead. Do not rebase only because `main` advanced; merge drift is advisory unless strict drift is explicitly enabled, while GitHub still blocks conflicts. Do not idle on `auto-response` or `check-docs`.
 - After GitHub throttling, check core quota before `scripts/pr prepare-run` or `merge-run`. A failed operation can retain its lock; verify no child remains, then recover only with its emitted token.
+- Local `scripts/pr`: unset `GITHUB_TOKEN`, `GH_TOKEN`, `HOMEBREW_GITHUB_API_TOKEN`; ambient tokens can select an exhausted or wrong identity.
 - Non-main PRs: do not run `scripts/pr prepare-run` or `merge-run`; they diff against `main`. Use review artifacts, exact base-head CI, revalidate `headRefOid`, then `gh pr merge --match-head-commit <verified-sha>`.
 - Merge guard shells: start `set -euo pipefail`; a failed `[[ ... ]]` alone does not stop a later merge command.
 - After `scripts/pr merge-run` removes its worktree, `cd` to a persistent repo before follow-up commands.
diff --git a/src/gateway/server-methods/agent.test-harness.ts b/src/gateway/server-methods/agent.test-harness.ts
index 048d466bcd78..22db85d9d396 100644
--- a/src/gateway/server-methods/agent.test-harness.ts
+++ b/src/gateway/server-methods/agent.test-harness.ts
@@ -437,20 +437,22 @@ export async function flushScheduledDispatchStep() {
   await Promise.resolve();
 }
 
-export async function waitForAcceptedRunDispatch(respond: ReturnType<typeof vi.fn>) {
+export async function waitForAcceptedRunDispatch(params: {
+  respond: ReturnType<typeof vi.fn>;
+  commandCallCount: number;
+}) {
+  const { respond } = params;
   const accepted = respond.mock.calls.some(([ok, payload]) => {
     return ok === true && (payload as { status?: string } | undefined)?.status === "accepted";
   });
   if (!accepted) {
     return;
   }
-
-  const commandCallCount = mocks.agentCommand.mock.calls.length;
   const respondCallCount = respond.mock.calls.length;
   for (let attempt = 0; attempt < 50; attempt++) {
     await flushScheduledDispatchStep();
     if (
-      mocks.agentCommand.mock.calls.length > commandCallCount ||
+      mocks.agentCommand.mock.calls.length > params.commandCallCount ||
       respond.mock.calls.length > respondCallCount
     ) {
       return;
@@ -863,6 +865,7 @@ export async function invokeAgent(
   },
 ) {
   const respond = options?.respond ?? vi.fn();
+  const commandCallCount = mocks.agentCommand.mock.calls.length;
   // Most cases only need to cross the accepted-ack timer; keep tests that own
   // timer semantics on their explicit clock while avoiding a real sleep here.
   const ownsDispatchTimers = options?.flushDispatch !== false && !vi.isFakeTimers();
@@ -882,7 +885,7 @@ export async function invokeAgent(
       },
     );
     if (options?.flushDispatch !== false) {
-      await waitForAcceptedRunDispatch(respond);
+      await waitForAcceptedRunDispatch({ respond, commandCallCount });
     }
   } finally {
     if (ownsDispatchTimers) {