Coding standards, frameworks, refactoring, and code review (16.6K skills)
Compare major AI coding tools and consult practical guidance on prompts, project instructions, debugging, workflows, and costs. Use the reference to choose suitable tools and identify why an AI-assisted coding task is stuck.
Guide development from problem definition and planning through testing, debugging, review, and final verification. It uses test-driven development, writing a failing test before implementation, and requires evidence before reporting work as complete.
Plan, run, and merge several small development tasks as one batch in Pi. A coordinator manages shared files and worktrees, while options control concurrency, add research and validation, or resume eligible tasks from an existing batch.
Move verified development work toward a merged pull request, a request to review and combine code changes. Push the branch, generate the request body, optionally trigger a review, and track progress through the merge.
Configure GSD through an interactive questionnaire. Choose a model profile and turn workflow features such as research, verification, and code review on or off, then merge the choices into the existing configuration.
Test an app idea through focused experiments and collect verified findings for development. It can also suggest what to explore next based on existing experiments and package findings as a reusable project skill.
Draft a GSD phase plan in Claude Code's ultraplan cloud while keeping your terminal free. Review and comment on the plan in a browser, then import it into your local workflow.
Undo development changes recorded by GSD using Git revert. Select recent commits or target an entire phase or plan, with dependency checks and confirmation before the rollback runs.
Clarify what a development phase should deliver before discussing implementation. Through structured questions and ambiguity scores, the skill produces SPEC.md with testable requirements, explicit scope boundaries, and acceptance criteria for judging completion.
Check for GSD updates, preview the changelog, and install the selected release after confirmation. Additional options synchronize managed skills across runtime environments, restore local customizations through comparison, or select a release candidate for installation.
Complete small development tasks with a plan, focused commits, and recorded progress. Add discussion, research, or verification when needed, and use task listings and status checks to find and resume unfinished work.
Ask external AI command-line tools to independently review a development phase plan. Collect feedback by reviewer in REVIEWS.md, then feed the results into GSD planning to support revisions.
Create and manage isolated GSD workspaces with repository copies and separate planning folders. Use one command to create an environment, review active workspace status, or remove a workspace and clean up its worktrees.
View project statistics covering phase progress, plan execution, completed requirements, Git history, and the project timeline. Use the combined view to check how development is progressing across plans, requirements, and recorded changes.
Review development plans with external AI tools, then revise and review them again while actionable concerns remain. Choose reviewers and a cycle limit to work toward a plan with resolved review findings.
Start a new development milestone for an existing project. The skill gathers goals, updates project documents, optionally researches new features, and creates scoped requirements and a phased roadmap for the next cycle of work.
Build and search a project knowledge graph, a map of information and its relationships. Check matching entries, relationship confidence, graph freshness, and changes since the previous build to understand the project's recorded structure.
Read the current project and workflow state, then show a short menu of suitable next actions. After a selection, hand off to one existing GSD command to carry out the work.
Run remaining GSD project phases through discussion, planning, and execution, while recording progress and phase documents. You can limit the phase range or keep decisions interactive; after all phases finish, the workflow audits and closes the milestone.
Extract decisions, lessons, recurring patterns, and unexpected findings from the artifacts of a completed development phase. It organizes these learnings into a LEARNINGS.md file using planning, summary, verification, acceptance testing, and state documents.
Handle small development tasks directly in the current session, such as typo fixes, configuration changes, and minor refactoring. It skips subagents and PLAN.md files for work that can be described briefly and finished quickly.
Routes GSD project management requests to the appropriate skill for configuration, workspace management, parallel work, pausing and resuming tasks, shipping completed work, and maintenance. It also covers inbox processing, project statistics, and command guidance.
Import an external plan into GSD after checking it against recorded project decisions, then validate the resulting plan file. It also supports converting a GSD-2 project back to the GSD v1 planning format.
Execute a development phase by grouping plans into waves, sets of tasks that can run together based on dependencies. Collect results, verify completion, and update progress, with options to run a selected wave or work sequentially with checkpoints.