agentsclimarketplace

Create pr

Skill getpipher/skills/skills/git-tools/skills/create-pr

Monorepo of @getpipher/* pi-packages and framework-agnostic Agent Skills — git workflows, quality, workspace, Solana, infra playbooks.

Install
npx -y skills add getpipher/skills --skill create-pr

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Create a pull request (GitHub) or merge request (GitLab) from FROM branch to TO branch

SKILL.md

3.5 KB, 770 tokens by cl100k_base, as published. Nobody here has run it

Bismillah! I'll create a pull request/merge request with the provided arguments: $ARGUMENTS

I need to parse the arguments as: FROM_BRANCH TO_BRANCH [--github|--gitlab]

Platform Selection

Flags:

  • --github - Create PR on GitHub using gh CLI (default)
  • --gitlab - Create MR on GitLab using glab CLI

Default behavior: GitHub (if no flag specified)

Platform Detection Logic

# Parse platform flag from arguments
if [[ "$ARGUMENTS" == *"--gitlab"* ]]; then
    PLATFORM="gitlab"
    CLI_TOOL="glab"
    PR_TYPE="merge request"
elif [[ "$ARGUMENTS" == *"--github"* ]]; then
    PLATFORM="github"
    CLI_TOOL="gh"
    PR_TYPE="pull request"
else
    # Default to GitHub
    PLATFORM="github"
    CLI_TOOL="gh"
    PR_TYPE="pull request"
fi

echo "📌 Platform: $PLATFORM ($PR_TYPE)"

Using with Thinking Mode

When executing this command with mode: think, I will:

  • Analyze the commit history between FROM_BRANCH and TO_BRANCH for context
  • Review all changed files to understand the scope and impact of modifications
  • Craft a comprehensive PR title that accurately reflects the changes
  • Generate detailed PR description with:
    • Clear summary of what changed and why
    • Impact analysis and potential risks
    • Testing considerations and verification steps
    • Related issues or dependencies
  • Consider if the PR scope is appropriate or should be split
  • Validate that the target branch choice is optimal

This thoughtful approach ensures high-quality PRs with better documentation and reviewer context.

Let me:

  1. Parse the arguments and validate input

    • Extract FROM_BRANCH and TO_BRANCH from the arguments
    • Validate that both arguments are provided
  2. Check repository status and prerequisites

    • Verify we're in a git repository
    • Check if the appropriate CLI is installed and authenticated:
      • GitHub: gh auth status
      • GitLab: glab auth status
    • Confirm both branches exist (locally or remotely)
  3. Prepare branches for PR creation

    • Fetch latest changes from remote
    • Switch to FROM_BRANCH and ensure it's up to date
    • Push FROM_BRANCH to remote if needed
  4. Create the pull request / merge request

    For GitHub (default):

    gh pr create --base TO_BRANCH --head FROM_BRANCH --title "..." --body "..."
    

    For GitLab:

    glab mr create --source-branch FROM_BRANCH --target-branch TO_BRANCH --title "..." --description "..."
    
    • Generate meaningful PR/MR title based on recent commits
    • Create comprehensive description with changes summary
    • IMPORTANT: Mention '@claude' in the PR/MR description and request review with specific focus areas:
      • Critical code changes and their impact
      • Architecture decisions and patterns
      • Security considerations
      • Performance implications
      • Testing coverage and edge cases
    • Provide context about the changes and areas of concern
    • NEVER include AI attribution, credits, or "Generated with assistance" footers
    • Return the PR/MR URL

Usage Examples

# Create PR on GitHub (default)
/git:create-pr dev main

# Create PR on GitHub (explicit)
/git:create-pr dev main --github

# Create MR on GitLab
/git:create-pr dev main --gitlab

Let me execute these steps now.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.