개발 스킬

Investigate bugs by clarifying expected behavior, reproducing the problem, and testing possible causes. The skill helps prepare a fix plan with risks and validation steps before code changes begin.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reproducing a bug and documenting its environment
• Investigating a regression or failing test
• Triage of production incidents or error spikes

Analyze existing code for complexity, duplication, coupling, and unnecessary layers, then propose prioritized simplifications that improve readability and maintainability. The workflow checks whether code can be removed or replaced with existing platform features before suggesting changes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Simplifying difficult-to-read code
• Reducing duplication and unnecessary abstractions
• Planning a maintainability refactor

Turn a long AI coding session into a structured handoff or resume artifact in Markdown or JSON. The Jev-backed CLI extracts useful continuation facts and possible memory candidates; review its output before relying on its claims.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Preparing a handoff after a long coding session
• Resuming work after a session has gone stale
• Extracting durable facts from a debugging run

Carry out planned development tasks and check whether the code matches the feature’s design and requirements. The workflow also keeps implementation and testing documents current, tracks task progress, and summarizes deviations, gaps, and follow-up work.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implementing tasks from a feature plan
• Checking code against design and requirements
• Updating implementation and testing documents

Review a feature before pushing by comparing its code with the design, tracing affected callers, and checking contracts, edge cases, security, and integration risks. Get findings ordered by severity with file references and recommended fixes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Review a completed feature before pushing
• Check whether code follows the approved design
• Find integration risks across changed files

Understand a code entry point by examining its purpose, behavior, and dependencies up to three levels deep. The skill produces a structured Markdown knowledge document, with an optional HTML overview when requested.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Document how a function or API works
• Map a module’s dependencies and execution flow
• Create a knowledge document for a folder

Clarify a feature’s problem, users, scope, success criteria, and open questions, then prepare its requirements, design, and testing documents. It helps teams compare approaches and review requirements before moving into design work.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Capturing a new feature idea and its scope
• Clarifying open product or UX questions
• Reviewing requirements before design work

Check whether AI DevKit is ready on the local machine, identify setup gaps it can address, and run the narrowest relevant setup for supported agent harnesses. Recheck status afterward and report remaining manual actions.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check whether local AI DevKit setup is ready
• Set up missing supported agent integrations
• Review remaining setup steps after onboarding

Review a feature design against its requirements. Check architecture, components, data models, API contracts, trade-offs, and non-functional needs, then clarify decisions and record reusable architecture choices before implementation begins.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check a design document against feature requirements
• Compare architecture options and trade-offs
• Find missing decisions before implementation

Update the top Unreleased section of CHANGELOG.md using commits since the latest release. Add one concise linked line per commit, using a pull request link when evidence is clear and a commit link otherwise, while preserving the file’s existing style.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding recent commits to an Unreleased changelog
• Creating release notes from commits since the last release
• Linking changelog entries to their commits or pull requests

Turn feature requirements, design notes, and test scenarios into an ordered implementation plan. Track task status, dependencies, blockers, scope changes, and next steps in the configured planning documents.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Breaking feature documents into implementation tasks
• Updating task status after completed work
• Recording blockers and newly discovered scope

Prepare or resume an isolated workspace for feature work by selecting and verifying a branch and Git worktree, then bootstrapping dependencies. It reports the active work directory and any workspace risks or skipped setup steps.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Set up an isolated workspace for a feature
• Resume work in an existing feature worktree
• Verify the branch and work directory before coding

Track development and debugging work with the ai-devkit task CLI. Record phases, progress, next steps, blockers, ownership, and fresh validation evidence so later readers can see what changed, what remains, and what supports the result.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Recording progress across development phases
• Tracking a debugging hypothesis and its next step
• Logging a blocker and marking it resolved

Publish a committed, reviewed feature branch by fetching and rebasing it onto the latest base branch, pushing safely, and opening or updating a review request on a supported Git host. Report the URL, commit, validation, and risks.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Push a reviewed feature branch for feedback
• Update a pull request after syncing with the base branch
• Open a merge request on the project's Git host

Coordinate a feature’s development phases, from requirements and design through planning, implementation, testing, and review. It checks project setup, routes work to the relevant phase skill, and uses documented progress and verification to guide what comes next.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Choosing the next phase for an existing feature
• Coordinating a feature from requirements through review
• Resuming feature work using its current documents and workspace

Prepare a focused Git commit by reviewing repository changes, staging only files related to the requested work, and choosing a concise conventional commit message. It also reports validation and any remaining unstaged files.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Creating a commit for a finished change
• Staging only files related to the requested work
• Preparing a PR-ready checkpoint

Explore a local map of code symbols and their relationships to answer structural questions about a repository. Find definitions, callers, imports, and nearby dependencies, then inspect the source files for details. The graph refreshes automatically after agent turns.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Finding callers or users of a function
• Checking a file’s imports and dependencies
• Locating a class or function definition

Explore an automatically refreshed code graph to locate symbols and inspect calls, imports, inheritance, and dependencies. Use it to trace flows, identify likely change impact, or map repository subsystems, then verify findings in the source files.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Find where a function or class is defined
• See which code calls or imports a symbol
• Trace dependencies between two symbols

Browse an automatically refreshed, AST-derived map of a repository through the Deeplake mount. Search symbols, inspect file neighborhoods, trace dependency paths, and estimate change impact, then read the actual source for confirmation. Supports TypeScript, JavaScript, and Python.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Finding where a symbol is defined
• Tracing callers, imports, or dependency paths
• Exploring repository subsystems

Investigate a slow sglang-omni stage by pairing a stack-enabled mapping trace with a production-like formal trace. The analysis attributes GPU kernel time to Python source lines and separates code location from the time users actually experience.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Finding Python code responsible for GPU time in an omni stage
• Comparing code attribution with production-like timing
• Reading kernel, overlap, and fusion analysis tables

Run configured evaluation benchmarks on the detected GPU hardware and update matching reference-table cells with the results. New hardware or workload combinations get new rows, while personal Local-* tables are skipped. The refresh is committed locally.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Refreshing benchmark reference results after running the suite
• Adding results for a new GPU or workload combination
• Rerunning selected benchmarks or failed rows

Plan a bounded, five-layer profiling run for one model and prepare the executor prompt. The workflow pauses for human approval before GPU work, verifies the agent’s completion, and routes evidence to local run artifacts and the shared methodology document.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Planning a profiling run for one model
• Scoping a resume run around missing profiling layers
• Reviewing a GPU profiling plan before execution

Set up application registration and authentication with Microsoft Entra ID. Guides API permissions, credentials, and OAuth 2.0 flows, with MSAL authentication library examples for user sign-in and service-to-service access.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Registering an application in Microsoft Entra ID
• Adding user sign-in to a console application
• Configuring API permissions for an application

Commit local changes to the current Git branch with a message based on the project's conventions. When staged and unstaged changes coexist, choose which to include. The skill does not push changes or create pull requests.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Saving local work as a commit
• Committing only changes already staged
• Staging and committing all local changes