Default Workflow
Developmentdefault-workflowDefault Workflow is an Agent Skill for development published in the GoogleCloudPlatform/magic-modules repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Fallback workflow for general implementation and debugging tasks that do not involve creating a new resource.
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/GoogleCloudPlatform/magic-modules/blob/HEAD/.agents/skills/workflows/default/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/default-workflow, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
default-workflow
This document outlines the structured 5-step lifecycle for formal implementation and debugging tasks in Magic Modules.
Execution Steps
1. Repo Sync
- Execute the
repo-syncskill (located in.agents/skills/operations/repo-sync/). This skill handles checking the sync status and prompting for action if needed.
2. Triage
- Gather context on the change or bug. Plan the change (New feature or bug fix) within schema or logic.
- Execute the
triageskill (located in.agents/skills/operations/triage/) to perform this work. - Transfers to Step 3: Approved implementation plan and file paths file.
3. Generate
- Once the first pass at the change is made, execute the code generation using the
generate-providerskill (located in.agents/skills/operations/generate-provider/) to compile machine code in the downstream. - Transfers to Step 4: Compiled provider binary.
4. Test and Debug
- Invoke the specialized
qa-test-runnersubagent using theinvoke_subagenttool to reproduce failures and interpret logs. The subagent evaluates if the test fails/passes and returns a human-readable interpretation of the results. - Transfers to Step 5: Human-readable Markdown report explaining whether the test succeeded or failed, and what discrepancy was found.
5. Fix
- This is a Remediation Planning step (similar to Triage). Take the results from
qa-test-runnerand compare against reference guides or user suggestions. Propose a specific fix code change to the user. - Execute the
fixskill (located in.agents/skills/operations/fix/) to perform this planning.
The Loop
Repeat steps 2-5 as needed during the session until the primary task is complete. Reset to Step 3 (Generate) after applying an approved fix to compile it!