fix: prevent duplicate [aw] No-Op Runs issues on transient search failure

[Copilot]

A transient API error (rate limit, network blip) during the issue search in ensureAgentRunsIssue() was silently swallowed, causing execution to fall through to issues.create — potentially producing duplicate [aw] No-Op Runs tracking issues when one already existed.

Changes

• handle_noop_message.cjs — replace core.warning + fall-through in the search catch block with a throw, so a failed search aborts rather than creates:

// Before: swallowed error, fell through to create
} catch (error) {
  core.warning(`Error searching for no-op runs issue: ${getErrorMessage(error)}`);
}
// create path always reached on both "not found" and "search error"

// After: propagates error to caller
} catch (error) {
  throw new Error(`Failed to search for existing no-op runs issue: ${getErrorMessage(error)}`);
}
// create path only reached when search succeeded with zero results

The main() caller already wraps ensureAgentRunsIssue() in a try/catch that logs a warning and returns early, so this change slots cleanly into existing error handling with no behavior change for the happy path.

• handle_noop_message.test.cjs — adds a test asserting that a search API rejection does not trigger issues.create or issues.createComment.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>handle_noop_message: transient search failure falls through to create and can produce duplicate [aw] No-Op Runs issues</issue_title>
<issue_description>## Context

Observed while running gh-aw v0.52.1 in production in samuelkahessay/prd-to-prod — a parallel agentic pipeline with concurrent agent work. Still present on current main at 6502173c8 (v0.53.1-2-g6502173c8).

Location

actions/setup/js/handle_noop_message.cjs — lines 24–63, ensureAgentRunsIssue()

Problem

The search for an existing [aw] No-Op Runs issue catches errors and falls through to the create path without distinguishing "search returned zero results" from "search threw an error":

// Lines 24-41
try {
  const { data } = await github.rest.search.issuesAndPullRequests({
    q: searchQuery,
    per_page: 1,
  });

  if (data.total_count > 0) {
    // ... return existing issue
  }
} catch (error) {
  core.warning(`Error searching for no-op runs issue: ${getErrorMessage(error)}`);
}

// Falls through to create on BOTH "not found" AND "search failed"
const { data: newIssue } = await github.rest.issues.create({ ... });

A transient API failure (rate limit, network blip) can create a duplicate [aw] No-Op Runs tracking issue when one already exists. Over time this fragments the no-op run history across multiple issues.

The caller already expects ensureAgentRunsIssue() to throw and abort posting if issue setup fails (main() lines 133–138), so treating a search failure as "not found" is the wrong behavior here.

Proposed fix

Distinguish between "search returned zero results" and "search threw an error." On search error, either retry or propagate a search-specific error rather than falling through to create:

let existingIssue = null;
try {
  const { data } = await github.rest.search.issuesAndPullRequests({ ... });
  if (data.total_count > 0) {
    existingIssue = data.items[0];
  }
} catch (error) {
  throw new Error(
    `Failed to search for existing no-op runs issue: ${getErrorMessage(error)}`
  );
}

if (existingIssue) {
  return { number: existingIssue.number, node_id: existingIssue.node_id };
}

// Only create if search succeeded and returned zero results

Environment

• Observed in production: v0.52.1
• Still present on: main at 6502173c8 (v0.53.1-2-g6502173c8)
• Pipeline: Parallel agentic workflow (decompose → dispatch → implement → review → merge)
• Repo: samuelkahessay/prd-to-prod</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes handle_noop_message: transient search failure falls through to create and can produce duplicate [aw] No-Op Runs issues #19606

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[Copilot]

Pull request overview

Prevents duplicate [aw] No-Op Runs issues by aborting issue creation when the “search for existing issue” call fails (e.g., transient API/rate-limit/network errors), instead of silently falling through to the create path.

Changes:

• Update ensureAgentRunsIssue() to throw on search failures rather than warning and continuing to create.
• Add a regression test ensuring a rejected search does not call issues.create or issues.createComment.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
actions/setup/js/handle_noop_message.cjs	Propagates search failures so the caller can abort before creating a potentially-duplicate tracking issue.
actions/setup/js/handle_noop_message.test.cjs	Adds test coverage for “search throws” behavior to ensure no issue/comment is created.

Comments suppressed due to low confidence (1)

actions/setup/js/handle_noop_message.cjs:40

• When rethrowing the search failure, this wraps the original exception and loses its stack/metadata. Since this repo runs on modern Node versions, consider using the Error cause option (or otherwise preserving the original error) so logs/debugging retain the underlying error details.

    throw new Error(`Failed to search for existing no-op runs issue: ${getErrorMessage(error)}`);

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.