agentleFS
Cogni-AI-OUCogni-AI-OU/cogni-ai-agent-skills/gh-aw-new/SKILL.md

gh-aw-new

Skill4 starsChanged 6 months agoMITView on GitHub

Create new GitHub Agentic Workflows (gh-aw) from scratch using the CLI extension. You MUST load this skill when creating new Agentic Workflow files.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 6 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create new GitHub Agentic Workflows (gh-aw) by installing the CLI extension and fetching official step-by-step creation instructions.

    line 12
  2. Very specific:

    When creating standard GitHub Actions (.yml) that don't use the Agentic Workflows framework.

    line 17
  3. Distinctive:

    If you simply need to execute an agent prompt in the CLI rather than scaffolding a repository integration.

    line 18
  4. Very specific:

    If not installed or needs upgrade: gh extension install github/gh-aw or gh extension upgrade github/gh-aw

    line 30
  5. Very specific:

    Use webfetch to retrieve the relevant instruction file based on the user's request, and read ALL instructions before proceeding:

    line 32
  6. Very specific:

    Execute the Action: Follow the fetched prompt's instructions exactly.

    line 35
  7. Distinctive:

    Review Changes: Run git status. Ensure the following are present:

    line 36
  8. Very specific:

    Update .gitattributes: Ensure .gitattributes contains: .github/workflows/*.lock.yml linguist-generated=true merge=ours.

    line 41
  9. Distinctive:

    Validate: Run repo-standard pre-commit hooks to ensure the new workflow meets quality and security standards:

    line 42
  10. Very specific:

    Commit and Push: Commit the workflow files, lock files, agent files, and .gitattributes.

    line 46
  11. Very specific:

    Dynamic Loading: Never guess the structure of a new workflow.

    line 50
  12. Distinctive:

    Artifact Inclusion: Always include regenerated lock files and agent definitions in the commit to ensure the workflow is ready for execution.

    line 51
  13. Distinctive:

    When the user wants to create a new workflow from scratch, add automation, or design a workflow that doesn't exist yet.

    line 55
  14. Very specific:

    If gh aw commands fail, verify that GitHub CLI (gh) is authenticated and that the github/gh-aw extension is installed with gh extension …

    line 72
  15. Very specific:

    If a command is missing, ensure you have the latest version: gh extension upgrade github/gh-aw.

    line 73
  16. Very specific:

    Check .lock.yml files; workflows must be compiled successfully before they can run.

    line 74

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: gh-aw-new

<!-- markdownlint-disable MD013 MD023 MD031 MD032 -->

Create new GitHub Agentic Workflows (gh-aw) by installing the CLI extension and fetching official step-by-step creation instructions.

When Not to Use

  • To modify or debug an existing agentic workflow (use github-aw or gh-aw-troubleshooting instead).
  • When creating standard GitHub Actions (.yml) that don't use the Agentic Workflows framework.
  • If you simply need to execute an agent prompt in the CLI rather than scaffolding a repository integration.

Common Pitfalls

  • Ignoring the Prompt: Trying to create the .md workflow file manually without fetching the official create-agentic-workflow.md prompt, resulting in invalid frontmatter.
  • Forgetting Compilation: Scaffolding the workflow but failing to run gh aw compile, leaving the repository without an executable .lock.yml file.
  • Missing Hooks: Bypassing pre-commit run --all-files after generation, pushing a workflow that violates repository standards.

Core Process

  1. Install or Upgrade gh-aw:
    • Check installation: gh extension list
    • If not installed or needs upgrade: gh extension install github/gh-aw or gh extension upgrade github/gh-aw
  2. Fetch the Appropriate Prompt: Use webfetch to retrieve the relevant instruction file based on the user's request, and read ALL instructions before proceeding:
    • Create New Workflow: https://raw.githubusercontent.com/github/gh-aw/v0.74.3/.github/aw/create-agentic-workflow.md
    • Create Shared Workflow: https://raw.githubusercontent.com/github/gh-aw/v0.74.3/.github/aw/create-shared-agentic-workflow.md
  3. Execute the Action: Follow the fetched prompt's instructions exactly. Use gh aw new <workflow-name> to scaffold.
  4. Review Changes: Run git status. Ensure the following are present:
    • .github/workflows/<workflow-name>.md
    • .github/workflows/<workflow-name>.lock.yml
    • .github/agents/ (if new agents were created)
    • .github/aw/actions-lock.json (if new actions were pinned)
  5. Update .gitattributes: Ensure .gitattributes contains: .github/workflows/*.lock.yml linguist-generated=true merge=ours.
  6. Validate: Run repo-standard pre-commit hooks to ensure the new workflow meets quality and security standards:
    pre-commit run --all-files
  7. Commit and Push: Commit the workflow files, lock files, agent files, and .gitattributes.

Core Principles

  • Dynamic Loading: Never guess the structure of a new workflow. Always fetch and read the appropriate ROOT/.github/aw/create-*.md prompt first.
  • Artifact Inclusion: Always include regenerated lock files and agent definitions in the commit to ensure the workflow is ready for execution.

When to Use

  • When the user wants to create a new workflow from scratch, add automation, or design a workflow that doesn't exist yet.

Commands / Usage Patterns

# Verify installation
gh extension list | grep gh-aw

# Create a new workflow (follow fetched guidelines)
gh aw new <workflow-name>

# Compile the new workflow
gh aw compile <workflow-name>

Diagnostics and Troubleshooting

  • If gh aw commands fail, verify that GitHub CLI (gh) is authenticated and that the github/gh-aw extension is installed with gh extension list.
  • If a command is missing, ensure you have the latest version: gh extension upgrade github/gh-aw.
  • Check .lock.yml files; workflows must be compiled successfully before they can run.

Limitations

  • This skill relies on fetching external Markdown files from the gh-aw repository. Ensure webfetch is available.

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.