[aw] Local Mac Runner Qwen/Ollama Agent failed

### Workflow Failure

**Workflow:** [Local Mac Runner Qwen/Ollama Agent](https://github.com/githubnext/self-hosted-aw/blob/codex/local-mac-all-in-one-runner/.github/workflows/local-macrunner-qwenollama.md)
**Branch:** codex/local-mac-all-in-one-runner
**Run:** https://github.com/githubnext/self-hosted-aw/actions/runs/27784119989

> [!WARNING]
> **Invalid or Unsupported Model**: The agent failed because the configured model name is invalid, unknown, or unavailable for this engine/account.

This is a **configuration issue**, not a transient error — retrying will not help.

How to fix this

Specify a valid model for the selected engine in the workflow frontmatter:

```yaml
---
engine: copilot
model: gpt-5-mini
---
```

To find valid models, check your engine/provider documentation (for Copilot see [supported models](https://docs.github.com/en/copilot/using-github-copilot/using-github-copilot-in-the-command-line#supported-models)).

> [!WARNING]
> **Engine Failure**: The `codex` engine terminated unexpectedly.

**Last agent output:**
```
2026-06-18T19:33:55.653866Z WARN codex_file_watcher: failed to unwatch /tmp/gh-aw/mcp-config/skills/.system: No watch was found. about ["/tmp/gh-aw/mcp-config/skills/.system"]
[warning] awf-reflect models fetch for http://api-proxy:10000/v1/models failed (attempt 1/5): models fetch returned 503 for http://api-proxy:10000/v1/models
[info] Retry attempt 1/4 for awf-reflect models fetch for http://api-proxy:10000/v1/models after 250ms delay
[warning] awf-reflect models fetch for http://api-proxy:10000/v1/models failed (attempt 2/5): models fetch returned 503 for http://api-proxy:10000/v1/models
[info] Retry attempt 2/4 for awf-reflect models fetch for http://api-proxy:10000/v1/models after 500ms delay
[warning] awf-reflect models fetch for http://api-proxy:10000/v1/models failed (attempt 3/5): models fetch returned 503 for http://api-proxy:10000/v1/models
[info] Retry attempt 3/4 for awf-reflect models fetch for http://api-proxy:10000/v1/models after 1000ms delay
[warning] awf-reflect models fetch for http://api-proxy:10000/v1/models failed (attempt 4/5): models fetch returned 503 for http://api-proxy:10000/v1/models
[info] Retry attempt 4/4 for awf-reflect models fetch for http://api-proxy:10000/v1/models after 2000ms delay
[warning] awf-reflect models fetch for http://api-proxy:10000/v1/models failed after 4 retry attempts: models fetch returned 503 for http://api-proxy:10000/v1/models
```

### Action Required

**Assign this issue to an agent** to debug and fix the issue.

Debug with any coding agent

Use this prompt with any coding agent (GitHub Copilot, Claude, Gemini, etc.):

````
Debug the agentic workflow failure using https://raw.githubusercontent.com/github/gh-aw/main/debug.md

The failed workflow run is at https://github.com/githubnext/self-hosted-aw/actions/runs/27784119989
````

Manually invoke the agent

Debug this workflow failure using your favorite Agent CLI and the `agentic-workflows` prompt.

- Start your agent
- Load the `agentic-workflows` skill from `.github/skills/agentic-workflows/SKILL.md` or
- Type `debug the agentic workflow local-macrunner-qwenollama failure in https://github.com/githubnext/self-hosted-aw/actions/runs/27784119989`

> [!TIP]
>
> Stop reporting this workflow as a failure
>
> To stop a workflow from creating failure issues, set `report-failure-as-issue: false` in its frontmatter:
> ```yaml
> safe-outputs:
> report-failure-as-issue: false
> ```
>
>

> Generated from [Local Mac Runner Qwen/Ollama Agent](https://github.com/githubnext/self-hosted-aw/actions/runs/27784119989) · [◷](https://github.com/search?q=repo%3Agithubnext%2Fself-hosted-aw+is%3Aissue+%22gh-aw-workflow-id%3A+local-macrunner-qwenollama%22&type=issues)
> - [x] expires on Jun 25, 2026, 7:36 PM UTC

Start with the workflow at .github/workflows/local-macrunner-qwenollama.md and inspect run 27784119989, using .github/skills/agentic-workflows/SKILL.md or the referenced debug.md for guidance. Determine the configuration or engine failure causing the run to fail; done means the workflow completes successfully without the reported model or codex errors.

Written by the indexing model from the issue text.