개발 스킬

Guide code changes through strict test-driven development: write a failing test first, add only enough production code to pass it, then refactor while tests stay green. The skill also calls for checks and separate commits at each phase.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Starting a change with a failing test
• Implementing one acceptance criterion at a time
• Refactoring with the full test suite passing

Use this skill when a project idea or request is still unclear. It asks focused questions, surfaces assumptions, compares possible directions and trade-offs, then turns the chosen direction into clear requirements for research or implementation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Clarifying a vague feature request
• Comparing design options and trade-offs
• Aligning stakeholders on a direction

Use this skill to turn a task description into a structured software specification. It searches related code, resolves ambiguity with clarifying questions, breaks work into testable tasks, and refines acceptance criteria from review feedback.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Turning a rough feature request into a specification
• Finding related code and existing patterns
• Clarifying assumptions before implementation

Coordinate decisions and shared state across groups of AI agents using several consensus protocols. The skill covers Raft leader election, Byzantine fault tolerance for unreliable participants, Gossip state propagation, CRDT conflict-free merging, and configurable weighted voting thresholds.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Choosing a protocol for an agent swarm
• Synchronizing state across distributed agents
• Merging concurrent updates without conflicts

Manage Git work for GSD plans with one structured commit per task, separate commits for planning files, branch status checks, history and diff summaries, and annotated milestone tags. Keep code and planning changes organized with consistent references to phases and tasks.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Create a structured commit for a completed task
• Commit planning files separately from code
• Check branch status and summarize changes

Carry a specification through implementation, testing, review, and up to three improvement cycles. The skill checks quality against a threshold, organizes changes into atomic commits, and reports requirement completion, test coverage, quality, and remaining work.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implementing a feature from an approved specification
• Reviewing and improving code until it meets a quality threshold
• Tracking completed requirements, test coverage, and remaining work

Coordinate a parallel code review across architecture, security, performance, testing, code quality, and documentation. The review combines weighted scores and critical findings into an approval, change request, or rejection recommendation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing a feature after implementation
• Checking code before merging
• Getting a scored review across several quality areas

Map a codebase by summarizing its folders, languages, frameworks, entry points, dependencies, and test layout. The resulting overview gives agents project context for research, specifications, and code generation, including awareness of conventions and likely imports.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Understanding a project before researching a change
• Reviewing the structure after major project changes
• Providing project context for specifications

Use this skill to create process definitions for the babysitter event-sourced orchestration framework. It guides work from requirements gathering and research through documentation, backlog planning, JavaScript implementation, and a structure checklist based on SDK patterns.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Scaffolding a new development methodology process
• Creating a domain-specific process definition
• Planning phases, inputs, outputs, and review gates

Finds open GitHub issues assigned to a5c-agent, shows their details, and processes them one by one through the Babysitter workflow. It summarizes each result and creates a branch and pull request for the work.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Processing all open issues assigned to a5c-agent
• Reviewing issue titles, labels, and links before work
• Creating a branch and pull request for each issue

Create Git checkpoints before risky work or after quality checks, recording changed files and line counts. List checkpoints, preview the changes a restore would undo, and return the working state to a selected checkpoint.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Saving a rollback point before refactoring
• Recording a safe state after quality checks
• Previewing and restoring a prior checkpoint

Implement only the code needed to meet the stated requirements, following the project’s existing structure, naming, imports, and error-handling patterns. It supports focused feature work and fixes without adding unrequested features or premature optimization.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implementing the smallest change that meets a requirement
• Matching code to an existing project’s conventions
• Making a focused bug fix

Analyze a babysitter process run in an external public repository. The skill reviews run events and process design to identify successes, failures, breakpoint patterns, and actionable improvements, then writes a retrospective with evidence such as journal events and file references.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing a named repository's babysitter run
• Finding a public run to analyze
• Understanding how another team uses babysitter processes

Coordinates agent sessions from initialization through handoffs and resumption. It can restore a dead agent's session state and manage persistent identity for Polecat and Crew agents, with commands for attaching to coordination and passing context.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Starting an agent session with a role
• Handing work and context to another agent
• Reviving a dead or stuck session

Manage hooks for the current session: enable or disable them, check their status, and inspect execution time and output size. Performance alerts use colors to distinguish normal, slow, and longer-running hooks.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Temporarily disable a troublesome hook
• Check hook status and profiling data
• Review hook execution time and output size

Investigate software failures by collecting logs, stack traces, reproduction steps, and related Git history. The skill checks known bug patterns and assigns a confidence score to a proposed root cause, helping guide the next debugging step.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Trace an error from its logs and stack trace
• Compare a bug with recent Git changes
• Check known causes such as race conditions or config errors

Coordinates long-lived Crew agents and transient Polecat workers by assigning work through hooks, handing tasks between agents, and nudging stalled sessions. It also explains a three-level hook hierarchy and provides commands for listing, feeding, and reviving agents.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Assign work to persistent and short-lived agents
• Hand off tasks between agents and prompt stalled workers
• Set up shared, role-specific, and rig-specific hooks

Use this skill to add an adapter for a CLI-based coding agent harness. It guides capability review, adapter and session parsing implementation, authentication detection, hook and plugin wiring, tests, and documentation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding support for a new CLI coding agent
• Implementing session file parsing
• Checking authentication detection

Coordinate pending changes from multiple agents by checking for branch conflicts, resolving them with a chosen strategy, and merging changes in dependency order. Then verify the integrated work with tests, lint, and a build before it lands.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Integrate changes from several agents
• Resolve conflicts before merging branches
• Verify an integration with tests and a build

Coordinate related beads—Git-backed units of work—as convoys in the Gas Town multi-agent framework. Assign work to agents, track distributed progress, check completion, and merge the finished branches into the main workflow.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Breaking a goal into related work units
• Assigning beads to agents
• Tracking parallel agent work

Monitor active agents and long-running workflows through health checks and heartbeats. Detect unresponsive work, choose a recovery action such as restarting or reassigning an agent, and report findings with trend analysis.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Checking whether active agents are responsive
• Finding stalled work in a long-running workflow
• Restarting or reassigning an unresponsive agent

Turn recurring code patterns, design decisions, and debugging lessons into reusable agent skills. The workflow scores patterns for usefulness and reliability, organizes them, assigns versions, and prepares bundles with examples and import instructions for other projects.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Capture reusable lessons at the end of a development session
• Turn code review findings into documented patterns
• Preserve debugging strategies for future projects

Review code across security, quality, performance, and maintainability. It reports only issues assessed with at least 80% confidence, assigns severity levels, suggests concrete fixes, and produces a status summary with blocking information and issue counts.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing a feature before it is merged
• Checking a bug fix for regressions
• Finding security or performance issues

Use this skill to plan and review system or API architecture. It helps define component boundaries, data flow, service integrations, and scaling choices, with guidance on interfaces, dependencies, caching, and database design.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning the architecture for a new feature
• Reviewing component boundaries and dependencies
• Choosing an API or service communication pattern