This command takes a raw idea or prompt and systematically dissects it using structured brainstorming techniques inspired by the UNC Writing Center's brainstorming methodology. It analyzes the idea for feasibility, scope, and predicted value, extracts a hypothesis, and proposes high-level solutions. If the idea is not…
This command is used to instruct Claude to generate GitHub issues with a clean, actionable, well-structured format. It is optimized for creating consistent GitHub issues for engineering planning, with consistent formatting across feat, fix, chore, docs, and epic types.
This command helps PR authors respond to review comments efficiently. It fetches PR details and comments using GitHub CLI, enters plan mode to strategically address each comment, and executes the response plan with code changes and comment replies.
This command helps reviewers conduct comprehensive code reviews efficiently. It analyzes PR changes, generates review feedback, and can either output a markdown review file (silent mode) or post comments directly to the PR (interactive mode with plan approval).
Instructions for nimble-giant/ailloy, covering ailloy — ai agent instructions, behavioral contract, available commands, available skills and workflow blanks.
Instructions for nimble-giant/ailloy, a project described as: The package manager for AI instructions. Author, compile, package, and distribute reusable & agnostic AI workflow packages — like Helm, but for AI agents.
The package manager for AI instructions. Author, compile, package, and distribute reusable & agnostic AI workflow packages — like Helm, but for AI agents.
Plan first, build second. All implementation work begins with analysis and planning. Review requirements, evaluate feasibility, and present a plan for approval before writing any code. Implementation is a separate, deliberate act.
Seven-phase structured brainstorming that analyzes an idea for feasibility, scope, and value and saves a report — never generates code. Use when the user says "brainstorm X", "is this worth building?", "analyze this idea", "pressure-test this", or "help me think through this".
Create well-formatted GitHub issues with conventional-commit-style titles, labels, and project-board assignment. Use when the user says "create an issue", "file a ticket", or "open a GitHub issue".
Research the project and generate a holistic AGENTS.md file at the repo root. Use when the user says "init agents.md", "create AGENTS.md", or the project lacks agent instructions.
Open a new pull request or update an existing one on the current branch. Use when the user says "open a PR", "push this up as a PR", or "create a pull request".
Fetch all PR review comments, plan strategic responses, apply code changes, and post replies in one pass. Use when the user says "respond to PR comments", "address the review", or "reply to the feedback on PR 42".
Generate a comprehensive PR description from the branch diff against main. Use when the user says "write the PR description", "draft the PR body", or asks for PR summary content.
Conduct a full code review with silent (markdown report) or interactive (posts inline comments) mode. Supports focused reviews by category. Use when the user says "review this PR", "security-review PR 42", or "do a code review".
Pre-flight planning ceremony that evaluates work against INVEST criteria, generates BDD acceptance specs, and sketches architecture before any code is written. Use when the user says "preflight", "plan this", or "let's scope this work".
Regenerate the PR description for the current branch from scratch. Use when the user says "update the PR", "refresh the PR body", or "regenerate PR description".