Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

[aw] Prompt Optimization reported incomplete result

Open
#307 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
35/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Active
Tech stack
github-actions

Research direction

Start with the failed Prompt Optimization run and the debugging guidance at https://raw.githubusercontent.com/github/gh-aw/main/debug.md. Read .github/skills/agentic-workflows/SKILL.md, then investigate the workflow at .github/workflows/prompt-optimization.md and its Ollama connectivity from the agent sandbox. Done means the workflow can perform its evaluation and scoring steps without reporting an incomplete result.

Written by the indexing model from the issue text.

Description

agentic-workflows
Workflow Failure

Workflow: Prompt Optimization
Branch: main
Run: https://github.com/githubnext/gh-aw-wizard/actions/runs/36928315280

[!WARNING]
Task Could Not Be Completed: The agent reported that the task could not be performed due to an infrastructure or tool failure.

Reasons:

  • The Ollama OpenAI-compatible server ((127.0.0.1/redacted) that the task description says was already started was unreachable from the agent sandbox, blocking all --evaluate/--score steps required by the optimizer skill.

    No existing "Prompt optimization:" PR was found (checked via gh pr list), so the task proceeded to the hill-climbing loop, which requires reaching the Ollama server at (127.0.0.1/redacted)

Connectivity attempts (all failed):

  1. curl (127.0.0.1/redacted) — connection refused, retried 10x over ~100s.
  2. ollama binary / ollama list / ps aux — ollama CLI not installed in this sandbox and no ollama process visible.
  3. Tried alternate gateway/bridge IPs (172.30.0.1, 172.17.0.1, 172.30.0.20) — refused or unreachable.
  4. host.docker.internal — DNS resolution failed directly from the sandbox (awf-resolved-config.json shows it should be allow-listed and enableHostAccess=true, allowHostPorts includes 11434, but direct DNS lookup still failed: "Could not resolve host").
  5. Routed through the squid-proxy (172.30.0.10:3128) to host.docker.internal:11434 — squid resolved it to 172.17.0.1 but got ERR_CONNECT_FAIL / Connection refused (111), retried 6x over ~100s.
  6. Routed through squid-proxy directly to 172.17.0.1:11434 — squid returned HTTP 403 (ACL denial), retried 6x over ~100s.
  7. Confirmed squid proxy itself works for allow-listed domains (https://registry.npmjs.org returned 200), so the proxy path is functional but the Ollama server on the host is not reachable/listening from this container's network namespace.

No repository files were changed; .optimizer/ was never created. No PR, review comment, or review was submitted, per instructions to avoid manufacturing output without a working measurement harness.

Per the workflow's own guidance to avoid inspecting infrastructure internals (docker sockets, mount tables, container networking internals) and the 2-recovery-attempt retry limit, I stopped after these distinct connectivity approaches rather than continuing to probe the sandbox/firewall internals.

This is a structured incompletion signal (report_incomplete), not a real task outcome. Any other safe outputs emitted alongside this signal (e.g., comments) describe the failure state, not a completed review or action.

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/gh-aw-wizard/actions/runs/36928315280
Manually invoke the agent

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

[!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:

safe-outputs:
  report-failure-as-issue: false

Generated from Prompt Optimization · copilot · 72.9 AIC · ◷

Dominant language
JavaScript
Stars
5
Forks
1
Avg merge
1d 7h
Merged PRs (30d)
30

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from githubnext/gh-aw-wizard

All issues in githubnext/gh-aw-wizard

Similar issues

More JavaScript issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.