create-github-action · diff
git:20260321.b056f9f to git:20260328.04e9e1c
20 added, 45 removed. Audit A to A.
---
name: create-github-action
- description: >
- Scaffold a traditional deterministic GitHub Actions CI/CD workflow. Trigger with
- "setup github actions", "create a test workflow", "add a ci pipeline", "setup PR validation",
- or when you need a standard build, set, deploy, lint, release, or security scan pipeline.
- This is distinct from agentic workflows — no AI is involved at runtime.
+ description: Scaffold a deterministic GitHub Actions CI/CD workflow
+ argument-hint: "[workflow-type: test|build|deploy|lint|release|security]"
allowed-tools: Bash, Read, Write
---
- ## Dependencies
-
- This skill requires **Python 3.8+** and standard library only. No external packages needed.
-
- **To install this skill's dependencies:**
- ```bash
- pip-compile ./requirements.in
- pip install -r ./requirements.txt
- ```
-
- See `./requirements.txt` for the dependency lockfile (currently empty — standard library only).
-
- ---
- # GitHub Actions Scaffolder
-
- You are an expert CI/CD Pipeline Architect. Your job is to scaffold traditional GitHub Actions workflows (deterministic automations with no AI at runtime).
-
- Read `references/action-types.md` before starting to understand triggers, permissions, and common action versions.
+ Follow the `create-github-action` skill workflow to scaffold a traditional deterministic
+ GitHub Actions CI/CD workflow (no AI at runtime).
- ## Execution Flow
+ ## Inputs
- Execute these phases in order. Do not skip phases.
+ - `$ARGUMENTS` — optional workflow type or purpose (e.g. `test`, `build`, `deploy`,
+ `lint`, `release`, `security`). Omit to start with discovery.
- ### Phase 1: Guided Discovery
- Conduct a short interview to understand the pipeline requirements:
- 1. **Category**: What does this workflow need to accomplish? (Test, Build, Lint, Deploy, Release, Security, Maintenance)
- 2. **Platform/Language**: What is the technology stack? (Python, Node.js, Go, Docker, .NET, Generic)
- 3. **Triggers**: When should this fire? (`pull_request`, `push`, `workflow_dispatch`, `schedule`, `release`)
+ ## Steps
- Wait for the user's answers before generating any files.
+ 1. If `$ARGUMENTS` specifies a workflow type, use it to seed Phase 1 discovery
+ 2. Follow the create-github-action phased workflow: confirm trigger events, runner OS,
+ required secrets/environment variables, job steps, and caching strategy
+ 3. Generate the `.github/workflows/<name>.yml` file
+ 4. Report the workflow path and setup instructions (secrets to configure, badges, etc.)
- ### Phase 2: Action Scaffold
- Once parameters are approved, use bash execution to run the scaffold script:
+ ## Output
- ```bash
- python ${CLAUDE_PLUGIN_ROOT}/scripts/scaffold_github_action.py \
- --category [category] \
- --platform [platform] \
- --triggers [triggers-list]
- ```
+ `.github/workflows/<name>.yml` with complete job definitions, trigger events, permissions,
+ and inline comments explaining each step.
- *Note: Ensure you pass the exact parameters determined in Phase 1.*
+ ## Edge Cases
- ### Phase 3: Post-Scaffold Instructions
- After successful execution, provide the user with the relevant next steps:
- 1. **Secrets Management**: If the workflow requires deploy keys or tokens (e.g., `PYPI_TOKEN`, `DOCKER_PASSWORD`), remind the user to add them to Repo Secrets.
- 2. **Review**: Advise them to review the generated `.yml` file in `.github/workflows/` before committing.
- 3. **Audit**: Offer to run `audit-plugin` to validate the YAML syntax.
+ - If `$ARGUMENTS` is empty: begin with workflow type discovery
+ - If the use case involves AI agents at runtime: redirect to `create-agentic-workflow`
+ - If secrets or environment variables are required: list them explicitly without values