fix(ui): retry the latest failed task attempt (#13765)

## Thinking Path

> - Paperclip manages work done by AI agents.
> - The task thread lets an operator retry a failed run.
> - Legacy runs with transcript output did not get a failure marker.
> - The thread could therefore offer Try again for an older failure.
> - The server correctly reused that failure's existing retry, even when
it had already failed.
> - This change keeps the latest failure actionable and reports stopped
retry responses to the operator.

## Linked Issues or Issue Description

**What happened?**

Try again could return success without starting work. An initial setup
failure had an empty transcript. Its later retry produced output and
failed. Only the initial failure had a retry marker, so the button kept
requesting the initial failure's already-failed successor.

**Expected behavior**

Try again targets the latest failed attempt. An already-stopped retry
response shows an error and refreshes the task's run state.

**Steps to reproduce**

1. Start a legacy adapter task that fails before producing transcript
output.
2. Retry it. Let this attempt produce output and a final failure comment
before it fails.
3. Click Try again in the task thread.
4. Before this fix, the click targets the original failure and replays
the stopped successor.

**Paperclip version or commit**

Reproduced against `1483bb8bcf`; the regression is also present on the
branch base `8813a50105`.

**Deployment mode**

Authenticated server with a legacy adapter. The bug is in the shared
task UI and retry API client.

Related: #11650 adds a different recovery-notice action. This change
fixes failed-run markers and retry response handling. Searches found no
duplicate of this failure case.

## What Changed

- Render legacy failure markers even when the run has a transcript or
final comment.
- Keep later cancelled automatic retries from replacing the failed run's
retry action.
- Reject already-stopped retry responses in the API client so existing
error feedback appears.
- Refresh task run queries after both successful and failed retry
requests.
- Add regression coverage for failed and timed-out attempts, execution
gates with output, and retry response states.

## Verification

- Before the fix, the new regression tests failed: two selected the
original failure, and four accepted a stopped successor as success.
- Targeted task-thread, retry API, marker, and issue-page tests: 279
passed.
- `pnpm --filter @paperclipai/ui typecheck`: passed.
- `pnpm --filter @paperclipai/ui build`: passed.
- `pnpm check:token-gates`: passed.
- Full UI suite: 6,529 tests passed across 626 files.
- [CI run
35650385023](https://github.com/paperclipai/paperclip/actions/runs/35650385023):
all 53 checks passed, including full workspace build, typecheck,
unit/integration suites, and browser tests. The redundant local
full-workspace test run was stopped after CI passed; it is not counted
as a completed local pass.
- Greptile: 5/5 on `608ee58c99`, with no review threads or unresolved
comments. The branch is mergeable.
- `pnpm -r typecheck` and `pnpm build` were attempted. Both stop at the
Runner's Rust checks because this host has no `cargo`. The full
workspace checks passed in CI.

## Risks

Low risk. This changes UI presentation and response handling only. The
server's exact-retry idempotency, authorization, execution ownership,
and recovery gates remain in place. No schema changes or live task
mutations. Existing documentation describes this retry action; the fix
restores that behavior.

## Model Used

OpenAI GPT-6 (Codex), with reasoning, repository tools, and code
execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass (targeted checks;
full-workspace limits described above)
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
(existing behavior restored; no documentation change needed)
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

Co-authored-by: Paperclip <noreply@paperclip.ing>

ui/src/api/agents.test.ts

@@ -0,0 +1,36 @@
+import { afterEach, describe, expect, it, vi } from "vitest";
+import { agentsApi } from "./agents";
+import { api } from "./client";
+
+afterEach(() => vi.restoreAllMocks());
+
+describe("agentsApi.retryFailedRun", () => {
+  it.each(["failed", "timed_out", "cancelled", "interrupted"])(
+    "reports a replayed %s retry instead of silently succeeding",
+    async (status) => {
+      const post = vi.spyOn(api, "post").mockResolvedValue({ id: "previous-retry", status });
+      await expect(agentsApi.retryFailedRun("agent-1", "original-failure", "company-1"))
+        .rejects.toThrow("The previous retry has already stopped");
+      expect(post).toHaveBeenCalledOnce();
+    },
+  );
+
+  it.each(["queued", "running", "succeeded"])("accepts a %s successor without dispatching again", async (status) => {
+    const post = vi.spyOn(api, "post").mockResolvedValue({ id: "successor", status });
+    await expect(agentsApi.retryFailedRun("agent-1", "original-failure", "company-1"))
+      .resolves.toEqual({ runId: "successor", issueId: null });
+    expect(post).toHaveBeenCalledOnce();
+  });
+
+  it("preserves a durable chat retry that is waiting for dispatch", async () => {
+    vi.spyOn(api, "post").mockResolvedValue({ actionId: "retry-action", status: "queued", runId: null, issueId: "issue-1" });
+    await expect(agentsApi.retryFailedRun("agent-1", "original-failure", "company-1"))
+      .resolves.toEqual({ runId: null, issueId: "issue-1" });
+  });
+
+  it("reports a skipped wake", async () => {
+    vi.spyOn(api, "post").mockResolvedValue({ skipped: true, message: "Task execution is paused." });
+    await expect(agentsApi.retryFailedRun("agent-1", "original-failure", "company-1"))
+      .rejects.toThrow("Task execution is paused.");
+  });
+});

ui/src/api/agents.ts

@@ -259,7 +259,16 @@ export const agentsApi = {
       reason: "retry_failed_run",
       failedRunId,
     });
-    if ("id" in result) return { runId: result.id, issueId: null };
+    if ("id" in result) {
+      // Exact retries are idempotent: a repeated click can return a successor
+      // that has already stopped. It does not mean a new attempt was queued.
+      if (["failed", "timed_out", "cancelled", "interrupted"].includes(result.status)) {
+        throw new Error(
+          "The previous retry has already stopped. Refresh the task and retry its latest failed run.",
+        );
+      }
+      return { runId: result.id, issueId: null };
+    }
     if ("actionId" in result) {
       if (result.status === "failed" || result.status === "cancelled") {
         throw new Error(

ui/src/components/TaskChatThread.test.tsx

@@ -2175,10 +2175,74 @@ describe("TaskChatThread runtime transcript selection", () => {
     },
   );
 
-  it.each(["active execution", "pending decision", "recovery hold"] as const)(
-    "does not promise legacy Retry during %s and restores it when the gate clears",
-    (gate) => {
+  it.each(["failed", "timed_out"] as const)(
+    "retries the latest legacy %s attempt even when it has a transcript and final comment",
+    async (status) => {
       const onRetryFailedRun = vi.fn();
+      transcriptState.transcriptByRun.set("latest-attempt", [{
+        kind: "assistant",
+        ts: "2026-08-25T18:01:01.000Z",
+        text: "Starting the requested revision.",
+      }]);
+      const run = {
+        runtimeMode: "legacy" as const,
+        adapterType: "claude_local",
+        agentId: "agent-1",
+        agentName: "Direct agent",
+        startedAt: null,
+      };
+      render(
+        <TaskChatThread
+          comments={[{
+            id: "failure-comment",
+            companyId: "company-1",
+            issueId: "issue-1",
+            authorAgentId: "agent-1",
+            authorUserId: null,
+            authorType: "agent",
+            presentation: null,
+            metadata: null,
+            body: "The startup handshake timed out.",
+            runId: "latest-attempt",
+            createdAt: new Date("2026-08-25T18:01:02.000Z"),
+            updatedAt: new Date("2026-08-25T18:01:02.000Z"),
+          }]}
+          onAdd={async () => {}}
+          issueStatus="todo"
+          onRetryFailedRun={onRetryFailedRun}
+          linkedRuns={[
+            { ...run, runId: "original-failure", status: "failed",
+              createdAt: "2026-08-25T18:00:00.000Z", startedAt: "2026-08-25T18:00:00.000Z",
+              finishedAt: "2026-08-25T18:00:02.000Z" },
+            { ...run, runId: "latest-attempt", status, errorCode: "acpx_handshake_timeout",
+              createdAt: "2026-08-25T18:01:00.000Z", startedAt: "2026-08-25T18:01:00.000Z",
+              finishedAt: "2026-08-25T18:01:02.000Z",
+              resultJson: { presentationDecision: { commentId: "failure-comment" } } },
+            { ...run, runId: "cancelled-automatic-retry", status: "cancelled",
+              errorCode: "execution_reconciliation_required",
+              createdAt: "2026-08-25T18:02:00.000Z", finishedAt: "2026-08-25T18:02:02.000Z" },
+          ]}
+        />,
+      );
+
+      expect(container.textContent).toContain("The startup handshake timed out.");
+      const buttons = container.querySelectorAll<HTMLButtonElement>('[data-testid="task-chat-run-failed-try-again"]');
+      expect(buttons).toHaveLength(1);
+      flushSync(() => buttons[0]!.click());
+      await Promise.resolve();
+      expect(onRetryFailedRun).toHaveBeenCalledExactlyOnceWith("latest-attempt");
+    },
+  );
+
+  it.each(["active execution", "pending decision", "recovery hold"].flatMap(
+    (gate) => [false, true].map((hasTranscript) => ({ gate, hasTranscript })),
+  ))(
+    "does not promise legacy Retry during $gate (transcript: $hasTranscript) and restores it when the gate clears",
+    ({ gate, hasTranscript }) => {
+      const onRetryFailedRun = vi.fn();
+      if (hasTranscript) transcriptState.transcriptByRun.set("legacy-failed", [{
+        kind: "assistant", ts: "2026-08-25T18:00:01.000Z", text: "Starting the task.",
+      }]);
       const failedRun = {
         runId: "legacy-failed",
         runtimeMode: "legacy" as const,

ui/src/components/TaskChatThread.tsx

@@ -1625,6 +1625,50 @@ export function TaskChatThread(props: TaskChatThreadProps) {
           },
         });
       }
+      // A failed run still needs its own retry target when it produced output
+      // or a final comment. Otherwise the thread keeps retrying an older run.
+      const sourceHasLegacyStop = !sourceIsPaperclipRunner && (
+        source.status === "failed" || source.status === "timed_out" ||
+        (source.status === "cancelled" && entries.length === 0)
+      );
+      if (sourceHasLegacyStop) {
+        settledRunIds.add(source.id);
+        const code = meta?.errorCode ?? "native_runner_process_exited";
+        const retryDetail = meta?.scheduledRetryAt
+          ? "Retry scheduled automatically."
+          : canRetryFailedRun
+            ? "You can retry this message now."
+            : "Your message is preserved.";
+        const aiRequest = interactions?.find((interaction) => interaction.kind === "connection_intent" && interaction.payload.purpose === "ai" && interaction.sourceRunId === source.id);
+        const detail = aiRequest
+          ? aiRequest.status === "pending"
+            ? "The selected AI account is unavailable. Fix it in the connection card."
+            : "This run stopped because its AI account was unavailable."
+          : source.status === "cancelled"
+            ? code === "execution_reconciliation_required"
+              ? "The previous execution must be checked before this task can continue. Your message is preserved. View the stopped run for details."
+              : "Execution was stopped before returning an answer."
+            : code === "provider_frame_too_large"
+            ? `Provider output exceeded the safe limit. ${retryDetail}`
+            : code.startsWith("workspace_git_scan_")
+            ? `Workspace setup failed before the agent started. ${retryDetail}`
+            : `The runner stopped before returning an answer (${code}). ${retryDetail}`;
+        const id = `${source.id}:failure`;
+        entriesWithFailures.push({
+          ms: toMs(meta?.finishedAt ?? meta?.startedAt ?? meta?.createdAt),
+          order: 3,
+          id,
+          item: {
+            id,
+            kind: "marker",
+            variant: "interrupted",
+            label: source.status === "cancelled" ? (meta?.startedAt ? "Stopped" : "Couldn't start") : "Run failed",
+            runId: source.status === "cancelled" ? undefined : source.id,
+            tone: source.status === "cancelled" ? "neutral" : "error",
+            detail,
+          },
+        });
+      }
       if (entries.length === 0) {
         if (sourceIsPaperclipRunner && sourceYielded) {
           settledRunIds.add(source.id);
@@ -1675,47 +1719,7 @@ export function TaskChatThread(props: TaskChatThreadProps) {
           });
           settledRunIds.add(source.id);
           settledReplyRunIds.add(source.id);
-        } else if (
-          !sourceIsPaperclipRunner &&
-          (source.status === "failed" || source.status === "timed_out" || source.status === "cancelled")
-        ) {
-          settledRunIds.add(source.id);
-          const code = meta?.errorCode ?? "native_runner_process_exited";
-          const retryDetail = meta?.scheduledRetryAt
-            ? "Retry scheduled automatically."
-            : canRetryFailedRun
-              ? "You can retry this message now."
-              : "Your message is preserved.";
-          const aiRequest = interactions?.find((interaction) => interaction.kind === "connection_intent" && interaction.payload.purpose === "ai" && interaction.sourceRunId === source.id);
-          const detail = aiRequest
-            ? aiRequest.status === "pending"
-              ? "The selected AI account is unavailable. Fix it in the connection card."
-              : "This run stopped because its AI account was unavailable."
-            : source.status === "cancelled"
-              ? code === "execution_reconciliation_required"
-                ? "The previous execution must be checked before this task can continue. Your message is preserved. View the stopped run for details."
-                : "Execution was stopped before returning an answer."
-              : code === "provider_frame_too_large"
-              ? `Provider output exceeded the safe limit. ${retryDetail}`
-              : code.startsWith("workspace_git_scan_")
-              ? `Workspace setup failed before the agent started. ${retryDetail}`
-              : `The runner stopped before returning an answer (${code}). ${retryDetail}`;
-          const id = `${source.id}:failure`;
-          entriesWithFailures.push({
-            ms: toMs(meta?.finishedAt ?? meta?.startedAt ?? meta?.createdAt),
-            order: 3,
-            id,
-            item: {
-              id,
-              kind: "marker",
-              variant: "interrupted",
-              label: source.status === "cancelled" ? (meta?.startedAt ? "Stopped" : "Couldn't start") : "Run failed",
-              runId: source.status === "cancelled" ? undefined : source.id,
-              tone: source.status === "cancelled" ? "neutral" : "error",
-              detail,
-            },
-          });
-        } else if (!sourceHasNativeStop && !lastCommentIdByRun.has(source.id)) {
+        } else if (!sourceHasNativeStop && !sourceHasLegacyStop && !lastCommentIdByRun.has(source.id)) {
           settledRunIds.add(source.id);
           const id = `${source.id}:terminal-notice`;
           entriesWithFailures.push({

ui/src/pages/IssueDetail.tsx

@@ -1562,6 +1562,8 @@ const IssueDetailChatTab = memo(function IssueDetailChatTab({
           tone: "success",
         });
       }
+    },
+    onSettled: () => {
       queryClient.invalidateQueries({
         queryKey: queryKeys.issues.runs(issueId),
       });