Coding standards, frameworks, refactoring, and code review (16.6K skills)
Debug programs with LLDB while keeping session state across commands. Set breakpoints, inspect variables and memory, and examine core dumps, with JSON results for agents and a debug adapter interface for editor clients.
Manage AI coding tool configurations through CC Switch from the terminal. Switch active providers, adjust settings, enable MCP servers that connect agents to external tools, and inspect API usage, costs, installed skills, and conversation sessions.
Design tools that AI agents can choose and use correctly. Define clear descriptions, inputs, and outputs, manage large search results, and plan approval checks, isolated execution environments, and interfaces for collaboration between agents.
Write small, reproducible Python scripts and run them in a sandbox using execute_python. Check the script's logic and output, then report the actual results or errors while clearly distinguishing executed results from code that has not been run.
Design the operating layer around a coding agent so it can read and edit code, run checks, and correct errors. Define completion criteria, permission boundaries, feedback, and recovery methods, and assess which tasks suit autonomous execution.
Plan how to add external capabilities to an AI agent using MCP, a connection protocol, or Skill Hub distribution. Compare integration methods, design tools and data access, and review context costs, credentials, and third-party security risks.
Design recovery paths for AI agents that loop, lose streamed responses, or hit service limits. Classify failures, choose retries or backup models, and set stopping rules to keep recovery attempts from causing further failures.
Design agents that operate desktop or browser software through screenshots, mouse actions, and keyboard input. Compare ways to locate interface elements, diagnose coordinate errors, handle changing screens, and evaluate task completion, operating speed, and safety boundaries.
Design agents that respond to incoming emails, external notifications, and timers. Plan how they handle interruptions, prioritize new requests, track concurrent tasks, and deliver results while longer operations continue in the background.
Create, update, or validate a concise architecture document from a specification, an idea, or existing code. It records shared design rules so separately developed parts follow consistent boundaries, dependencies, state changes, and data ownership.
Delegate a feature, bug fix, or substantial code change and carry it through implementation, review, and verification. Start from a task description or issue link, with options for the workflow route and review depth.
Review code changes with several independent reviewers working in parallel, then receive organized findings. Request a quick or thorough review, choose to hand off fixes, or run without interaction when another workflow or CI job initiates the review.
This entry stores the BMad Method module’s metadata rather than providing an executable workflow. Use the separate bmad skill with setup method to initialize the module or report its existing setup state.
Implement approved development stories and code changes with Amelia, a senior software engineer persona. Follow a test-first process: write a failing test, make it pass, then improve the code while checking every acceptance criterion.
Create and maintain repository instructions for AI agents in AGENTS.md. Check rules against project evidence, preserve existing requirements during revisions, and record observed mistakes so future agents receive concise, relevant guidance for working in the codebase.
This metadata entry records the Toolsmith module. It is not run directly; use the separate bmad skill to set up the module on first use or report its current state afterward.
Customize installed BMad skills by turning requested changes into TOML configuration files. Adjust exposed agent or workflow settings, separate team rules from personal preferences, and verify how the changes combine with existing settings.
Turn product requirements and user experience plans into technical architecture with Winston, a system architect persona. Discuss technology choices and their trade-offs, with an emphasis on established tools, developer productivity, and designs that can be implemented.
Run one iteration of a BMad unattended development loop, with optional route and review selections. The skill follows a generated workflow and reports a halt reason or command failure when execution cannot continue.
Build and maintain BMad skills, agents, and modules through a guided conversation. Turn prompts or session logs into reusable skills, review existing tools, and prepare modules for distribution, with approval before files are written.
This metadata record supports the BMad Core Tools module. Use the separate bmad skill to set up the module or report its current state; this record is not intended to be invoked directly.
Write an architecture decision record, a document explaining a team's design choice. Capture the context, rejected alternatives, and future costs so newcomers can understand the reasoning, then review the draft before saving it.
Review code changes or documents through configured review lenses, each representing a different perspective. Findings identify the location, triggering condition, suggested fix, and potential consequence, with reports available as JSON, Markdown, or both.
Guide a person through reviewing a commit, pull request, file, or directory. The skill generates and follows a BMad workflow to lead the review of the selected target.