GitHub Actions Workflow Failure Debugger

Find whether a failing CI job was never scheduled, could not start, or failed in a step. This practical engineering workflow asks for relevant evidence, spells out how to use it, and produces a reviewable output instead of generic suggestions or invented details.

GPT / Claude / Gemini6 variables
Prompt
You are a specialist assisting with github actions workflow failure debugger. Produce a practical, evidence-led result from the inputs below.

WORKFLOW YAML: {WORKFLOW_YAML}
RUN STATUS: {RUN_STATUS}
JOB LOGS: {JOB_LOGS}
TRIGGER EVENT: {TRIGGER_EVENT}
RUNNER ENV: {RUNNER_ENV}
RECENT CHANGES: {RECENT_CHANGES}

Task workflow:
1. Separate workflow trigger and runner provisioning failures from test failures.
2. Inspect permissions, environment, dependencies, secrets availability and cache.
3. Propose smallest verifiable YAML or script changes.
4. Never advise bypassing required safety checks..

Accuracy and safety rules:
- Only use facts present in my inputs or explicitly identified as checked references; distinguish facts from assumptions.
- If a required item is missing, mark it unknown and explain the minimum verification needed. Do not fabricate examples as if they were real.
- Missing runner logs cannot be treated as evidence that application tests failed.
- Explain important tradeoffs and any limitation caused by my stated format, platform, or audience.
- Avoid padding, repetitive claims, fake metrics, and promises of guaranteed ranking, performance, admission, or success.
- Clearly label ready-to-use material versus suggestions that still need verification.

Deliverables:
1. A failure classification
2. Evidence checklist
3. Minimal workflow patch
4. And safe rerun path.
5. Brief final checklist showing how I can verify the result against the provided inputs.

Before finalizing, check the work for contradictions, unsupported facts, and any missing edge cases. Prefer specific, useful instructions over generic advice.
Quick brief
Purpose

Find whether a failing CI job was never scheduled, could not start, or failed in a step.

Expected output

The response should contain a failure classification, evidence checklist, minimal workflow patch, and safe rerun path. The output should be formatted so it can be copied, checked, and used without guessing what was assumed.

Customize before copying

Replace these placeholders with your own context before you run the prompt.

{WORKFLOW_YAML}{RUN_STATUS}{JOB_LOGS}{TRIGGER_EVENT}{RUNNER_ENV}{RECENT_CHANGES}
Works well with
GPT
Claude
Gemini
Variations
Make the response shorter while keeping all required evidence and verification steps for GitHub Actions Workflow Failure Debugger.
Produce a beginner-friendly version that defines specialized terms and includes a concrete example for GitHub Actions Workflow Failure Debugger.
Use a professional, implementation-ready format with explicit edge cases, constraints, and acceptance criteria for GitHub Actions Workflow Failure Debugger.
What this prompt helps you do
Find whether a failing CI job was never scheduled, could not start, or failed in a step. This practical engineering workflow asks for relevant evidence, spells out how to use it, and produces a reviewable output instead of generic suggestions or invented details.
When to use it
Use this template when your immediate task is find whether a failing CI job was never scheduled, could not start, or failed in a step. It is most useful when you can provide the requested input fields and review the result against real examples.
How it works
The workflow anchors its reasoning in the supplied material, organizes the user's goal, and executes this task-specific process: Separate workflow trigger and runner provisioning failures from test failures; inspect permissions, environment, dependencies, secrets availability and cache; propose smallest verifiable YAML or script changes; never advise bypassing required safety checks.
Best practices
Include the failing workflow YAML, run URL or logs, and relevant event. Replace every input placeholder with genuine context, including examples of the expected result. Review the first draft against your own constraints before reusing it.
Common mistakes
Skipping key inputs, accepting unsupported claims, and publishing or executing generated output without review are frequent failure modes. In this task especially: Missing runner logs cannot be treated as evidence that application tests failed.
What you should expect back
The response should contain a failure classification, evidence checklist, minimal workflow patch, and safe rerun path. The output should be formatted so it can be copied, checked, and used without guessing what was assumed.
Limitations
Missing runner logs cannot be treated as evidence that application tests failed. This is an assisted drafting and analysis template: it cannot independently validate inaccessible files, live platform data, or unprovided sources.
Model notes
Works with GPT, Claude, and Gemini. Provide the stated inputs as text, with redacted private information where appropriate.
Real-world applications
Suitable for people using AI to find whether a failing CI job was never scheduled, could not start, or failed in a step in routine development workflows, with manual verification before reuse.
How to tell if it worked
The proposed fix has a measurable CI verification step and does not weaken controls. A second reviewer should be able to trace assertions to supplied inputs, and the final artifact should pass the verification checklist.
Where to go next
Explore related prompts in the development category when you need adjacent tasks, reviews, or follow-up work.