[agent-suggestions] Agent suggestions - 2026-06-08

[github-actions]

Agent Suggestions

Date: 2026-06-08

1. AW Failure Handoff Coordinator

Trigger: issues (opened, reopened) for titles matching [aw] * failed, plus optional daily schedule sweep for missed events.
Purpose: Automatically deduplicate recurring internal workflow-failure issues and convert them into a single actionable remediation thread per error signature (instead of repeated one-off failure tickets that require manual assignment).
Proposed safe outputs: add-comment (link duplicate to canonical tracker), create-issue (canonical recurring-failure tracker), noop (no recurring pattern).
Evidence:

• Repeated open [aw] ... failed issues with identical manual next step text: [aw] Code Duplication Detector failed #1213, [aw] Code Complexity Detector failed #1214, [aw] Bug Hunter failed #1205, [aw] Product Manager Impersonator failed #1204, [aw] Internal: Downstream Health failed #1203, [aw] Project Summary failed #1202, [aw] Internal: Agent Deep Dive failed #1224
• Existing recurring-pattern reports already identify repeated failure classes (safeoutputs/outputs.jsonl missing, auth/quota startup failures): [agent-efficiency] Recurring agent runs omit safe outputs, causing detector workflow failures #1209, [agent-efficiency] Recurring agent run failures from uv setup and copilot auth/quota handling (last 7 days) #1231
• Existing generic workflows do not perform this handoff/dedup orchestration:
  • issue-fixer (issues: opened) triages/fixes one issue at a time (gh-agent-workflows/issue-fixer/README.md)
  • pr-actions-fixer is manual workflow_dispatch with run ID (gh-agent-workflows/pr-actions-fixer/README.md)
  • duplicate-issue-detector comments on similarity but does not maintain canonical recurring failure trackers by normalized runtime signature (gh-agent-workflows/duplicate-issue-detector/README.md)

Why not covered already:
Current workflows either fix a single issue/PR or detect duplicates on open; none automatically cluster recurring [aw] operational failures into one persistent remediation issue and route future failures to it.

Duplicate Checks

• Open agent-suggestions issue reviewed: [agent-suggestions] Auto-escalate recurring downstream-health failures into deduplicated action issues #1198 (downstream-health escalation) — related domain but distinct scope.
• Open PR search reviewed for in-flight equivalent workflow additions: repo:elastic/ai-github-actions is:pr is:open ("issue-fixer" OR "pr-actions-fixer" OR "duplicate-issue-detector" OR "agentic-workflows").

Downstream Signals

• internal-downstream-health reports recurring downstream engine/auth failures that would also benefit from the same handoff model once applied in downstream repos: [downstream-health] Downstream health report for 2026-06-06 #1233.
• Related recurring downstream failure evidence referenced in [downstream-health] Downstream health report for 2026-06-06 #1233 (e.g., elastic/ai-github-actions-playground/issues/2988, elastic/ai-github-actions-playground/issues/2933).

Suggested Next Steps

• Define a normalized failure-signature schema (workflow + stable error fingerprint) for [aw] issues.
• Prototype aw-failure-handoff-coordinator as a workflow that upserts one canonical tracker per signature.
• Add links from new [aw] failure issues to existing canonical trackers instead of spawning parallel remediation threads.

Note

🔒 Integrity filter blocked 114 items

The following items were blocked because they don't meet the GitHub integrity level.

• #1128 search_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #1175 search_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #1175 list_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• #392 list_pull_requests: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• Mint Ephemeral Tokens #1067 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• Issue Fixer can use Ephemeral Token #1025 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [redis_otel] "Keys per Database" dashboard panel always renders "Avg TTL (ms)" as null integrations#19437 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.storage integrations#19431 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.compute integrations#19430 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.application integrations#19429 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.ticket integrations#19428 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.network integrations#19427 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.incident integrations#19426 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.identity integrations#19425 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.exposure integrations#19424 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• [LogsDB] [Stack 8.19.16-SNAPSHOT] [axonius] Failing test daily: system test: default in axonius.alert_finding integrations#19423 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• ... and 98 more items

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

---

What is this? | From workflow: Trigger Agent Suggestions

Give us feedback! React with 🚀 if perfect, 👍 if helpful, 👎 if not.

• expires on Jun 15, 2026, 1:53 PM UTC

[github-actions]

This issue was automatically closed because it expired on 2026-06-15T13:53:03.617Z.

Closed by Workflow