[aw] Failure Investigator (6h)

Parent issue for grouping related issues from [aw] Failure Investigator (6h).

Sub-issues are automatically linked below (max 64 per parent).

Workflow: [aw] Failure Investigator (6h)

• expires on Jul 24, 2026, 11:41 PM UTC-08:00

Additional untracked P1 gap found (2026-07-28 6h sweep) — no sub-issue created (create_issue quota reached this run)

Linter Miner — Copilot CLI subagent (task tool) fails instantly with "No model available"

• Workflow: .github/workflows/linter-miner.lock.yml
• Run: §30384667352 — failed 2026-07-28T18:04:30Z, job agent exit code 1
• Evidence: 4 separate task/subagent spawns (17:56:44Z, 17:59:01Z, 18:01:28Z, 18:04:19Z) all failed instantly with:

  [copilot-sdk-driver] [sdk-driver] error: Execution failed: Error: No model available. Check policy enablement under GitHub Settings > Copilot
  

• Distinct from [aw] [aw] Failure Investigator (6h) has no AI credits pricing for model (claude-opus-5) #48558 (AI-credits pricing gap for primary model claude-opus-5) — this is a policy-enablement failure for subagent dispatch specifically, not a primary-model pricing/catalog issue.
• Recommended fix: check whether Copilot subagent (task-tool) model access needs a separate policy/entitlement flag from the primary agent model, add a fail-fast guard when this error is seen instead of retrying silently for ~8 minutes, and file with Copilot support if the policy gap is persistent rather than transient.
• Please open a dedicated [aw-fix] P1 sub-issue for this and link it here if it recurs in the next 6h window.

Generated by 🔍 [aw] Failure Investigator (6h) · age00 · 140.3 AIC · ⌖ 26 AIC · ⊞ 5.2K · ◷