epam

Five-axis code review before merge, plus responsive-parity and documentation-accuracy gates. Use for quality passes after implementation, before merge, and when asked to review a diff.

Audit and fix untranslated user-visible text — titles, headings, button and menu labels, aria-label/title/placeholder/alt, empty states, errors, tooltips. In libs/* every string comes from a `*Labels` prop with an English default and the lib never imports…

Audit and fix colors and typography in libs/* so the host controls both through props. Colors come from a `*Colors` prop as `var(--<prefix>-x, fallback)` in the component's .module.scss, set with `buildCssVars`; typography comes from a `*ClassName` prop (or…

Review a diff or pull request for @epam/ai-dial-ui-kit adoption — hand-rolled markup the kit already provides (ErrorText, CaptionText, StateScreen, Notification, LinkButton, Dropdown, Spinner, …), legacy Dial* components, ErrorText/CaptionText a11y wiring,…

On-demand index of the AI DIAL Chat design docs in docs/. Use when you need ground-truth on how the app works or is built — overall architecture, technical/product requirements, the authentication subsystem (login, logout, OIDC, session cookies, encrypted…

Deep codebase refactoring audit for AI DIAL Chat. Collects metrics, compares with prior local plans, detects dead-code candidates, and writes or updates refactoring-backend.md, refactoring-frontend.md, and refactoring.md index docs (git-excluded). Use when…

Discover and fix CVEs in AI DIAL Chat dependencies or container images and backport fixes to the two newest release-X.Y branches. Automatically scan when no CVE or report is supplied. Use for vulnerability remediation with release backports, or to backport an…

Use this skill whenever the user wants to commit, push, or ship changes in a git repository. Triggers include: "commit changes", "push my changes", "ship it", "commit and push", "create a branch and commit", "make a PR", "open a pull request", "create draft…

Helps update UUI documentation, add doc examples, configure Property Explorer, and manage component API documentation. Use when adding documentation examples, updating Property Explorer configs, generating API references, working with UUI documentation site,…

Guides the UUI pull request process including branch naming, pre-PR checklist, changelog updates, and quality requirements. Use when preparing a pull request, writing commit messages, or following UUI PR requirements.

Guides the UUI package release process including stable and beta releases, changelog updates, and handling failed releases. Use when releasing UUI packages, updating changelog, or troubleshooting release issues. For maintainers only.

Helps create and modify UUI (EPAM Unified UI) components following established patterns. Use when creating new components, modifying existing components, working with withMods, component props, styling, or component architecture in the UUI library.

Helps work with UUI DataSources (ArrayDataSource, LazyDataSource, AsyncDataSource) powering PickerInput, DataTable, FiltersPanel, and other data-driven components. Use when implementing or fixing features that load, filter, sort, or display lists of data.

Helps create and maintain E2E and screenshot tests for UUI components using Playwright. Use when adding E2E tests, creating screenshot tests, updating preview configurations, or working with Property Explorer previews for testing.

Fetches GitHub issues from URLs and creates implementation plans for UUI. Use when the user provides a GitHub issue link (github.com/.../issues/N), asks to plan work from an issue, or implement/fix an issue by URL.

Helps work with UUI services layer including ContextProvider, ApiContext, ModalContext, NotificationContext, and useUuiContext. Use when implementing features that need modals, notifications, API calls, error handling, or routing.

Required before you commit, push or open a pull request here, asked or not. Holds the preconditions, the subject line and its trailers, the draft PR, and the issue label afterward.

Start here before the first edit of any file — a one-liner and a docs change included. Holds the steps a task passes through, the issue number and branch it needs, the tests, docs and UI test cases it owes, and where the branch ends up. The plan-change,…

Check, add, update or fix the UI test cases under docs/testcases/UI/. Required for every change that alters what a user sees or can do in the GUI — a client/ change, or an API or other backend change that reaches the GUI — and on its own whenever asked to…

Create, check, connect to, extend or remove a throw-away AWS sandbox for live testing (a Kubernetes 1.15 master on EC2, tagged to one developer, driven by scripts/sandbox.sh, and reached from local containers through a tunnel), run the platform's API, its…

Open a user-visible client/ change in a browser against a real Cloud Pipeline deployment, or a dev sandbox. Use when the diff changes UI layout, routing, client state or rendered data — not for tests, the test/ harness, lint/build config, or non-visual edits.…

Find and run the checks a change owes, from what it touches and what that can reach. Use after editing any file, when about to call a change finished, or when asked whether something passes.

Create a new GitHub issue in this repository — a bug report, an enhancement request, or a followup/sub-issue found while implementing another task. Use when asked to file, open, or write up an issue, not when implementing one that already exists.

Required before adding or changing anything an agent reads — a convention in AGENTS.md, a procedure as a skill, a pattern-scoped rule. Holds which of those a thing belongs in, the wiring each needs before it loads at all, and what to keep out of them.

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — typically a fresh CI-generated pre-release (e.g. `X.Y.Z-rc.N`) or a stable cut. Reads the auto-generated notes off the GitHub release, classifies and rewrites each…

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.

Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.

Five-axis code review before merge. Use for quality passes after implementation, before merge, and when asked to review a diff.

Design-to-code workflow for Figma designs. Use when the user shares a Figma URL or asks to implement a design into the codebase.

Five-axis code review before merge. Use for quality passes after implementation, before merge, and when asked to review a diff.

Use this skill whenever the user wants to commit, push, or ship changes in a git repository. This skill is intended to run with the developer's configured default Haiku model. Triggers include: "commit changes", "push my changes", "ship it", "commit and…

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — typically a fresh CI-generated pre-release (e.g. `0.16.1-rc.0`) or a stable cut. Reads the auto-generated notes off the GitHub release, classifies and rewrites each…

Find the one authoritative source for a question about this repo — a capability spec, an archived change's rationale, or an ops doc. Use before answering "what is this supposed to do", "why was it built this way", or any question about env vars, theming, the…

Use this skill whenever the user wants to commit, push, or ship changes in a git repository. Triggers include: "commit changes", "push my changes", "ship it", "commit and push", "create a branch and commit", "make a PR", "open a pull request", "create draft…

Read unresolved GitHub code review threads for the pull request associated with the current branch, classify each comment, and implement and verify required code fixes. Post inline replies only when the user explicitly asks to reply after the changes have…

Interactively create GitHub issues (Bug, Feature, Task) for the current repository. Infrastructure changes are Tasks auto-labeled `ops-request`. Parses optional args for type and description, asks targeted questions, assigns labels, and creates the issue via…

Verify an OpenSpec change's browser-observable acceptance scenarios against the live local app. Reads the change's delta + affected consolidated specs, builds a VerificationRequest, spawns the spec-verification-gate sub-agent (which drives the Playwright…

Use when the user is diagnosing a TimeBase failure from logs, stack traces, startup problems, runtime errors, or live operational symptoms such as stuck cursors, loaders, connections, or locks. Use this when the user pasted evidence and wants to know what it…

Use for TimeBase WebAdmin or Web Gateway REST API and STOMP integration: explain requests and authentication, build or debug custom clients, and work with streams, schemas, queries, stream imports and exports (including QSMSG), views, topics, playback,…

Use for TimeBase queries, views, stream schemas, and schema migrations. Generate, repair, review, or explain QQL and DDL, including creating or changing streams and views. Return small MCP-backed results directly, use the Python client skill when the task…

Use when user explicitly asks for TimeBase Java client code, for example typed stream reads/writes, stream creation, bound QQL execution from Java, client Maven/Gradle setup (Community or Enterprise Edition), or TimeBase Java client debugging.

Use when the user explicitly asks for TimeBase C++ client code, for example connecting to TimeBase from C++, reading/writing streams, dxapi library acquisition and CMake/vcxproj setup, or TimeBase C++ client debugging.

Use when the user explicitly asks for TimeBase C# or .NET client code, for example typed stream reads/writes, stream creation, bound QQL execution from C#, Deltix.Timebase package setup, or TimeBase .NET client debugging.

Use when the user needs TimeBase Python client code, for example scripts, apps, services, local file export, large result extraction, pandas workflows, visualization prep, or downstream Python computation. Do not use for pure MCP discovery or pure QQL tasks…

Use when the user asks to review, critique, validate, or approve a design doc under docs/designs/. Appends a Review Notes block; on explicit approval, strips notes and flips Status to Approved.

Use before creating a PR or claiming a change is ready (or on explicit user invocation). Self-reviews the current diff against the patterns this team's reviewers consistently flag.

Formats file and URL parameters for tool calls. You must analyze the target tool's parameter names and descriptions to choose the correct format (data URI, base64, text, or URL ref).

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — typically a fresh CI-generated pre-release (e.g. `0.8.0-rc.1`) or a stable cut. Reads the auto-generated notes off the GitHub release, classifies and rewrites each…

Use when the user asks to "create a PR", "open a pull request", "raise a PR", "ship this", or "/quickapps-create-pr" in this repo. Drives the project's end-to-end PR flow — feature branch, conventional-commit title/body, the pre-PR format+lint gate, push, and…

Use when the user wants to draft or file a GitHub feature-request issue for the Quick Apps 2.0 **editor** — the no-code app-builder UI that lives in the epam/ai-dial-chat repo ("create an issue for the quickapps editor", "draft a chat feature request for…

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — a fresh CI-generated pre-release (e.g. `0.13.0-rc.0`) or a stable cut. Reads the auto-generated notes off the GitHub release, tidies inconsistent phrasing and issue…

Structured CRITICAL→NIT review for diffs or paths; review-only until the user requests code changes.

End-to-end feature work across Next.js UI, Redux store, tests, and verification commands for this repo.

Align new edits with neighboring files—imports, Redux epics, Next.js patterns, and Tailwind rules in dial-mind-map-frontend.

Run the standard npm lint, test, format, and build checks expected before sharing dial-mind-map-frontend changes.

Use when the user wants an existing OpenSpec change implemented by sub-agents while this session stays a coordinator/reviewer — "coordinate the implementation", "coordinator mode", "implement <change> with subagents", "delegate this change". Assumes…

Use when the user wants to hand off the current branch's changes to a frontend developer/agent — "summarize changes for the FE", "write FE handoff", "what API changes does the frontend need". Works for a raw POC branch or one or more implemented opsx changes.…

Use when the user wants to generate a pull/merge request (PR/MR) summary for the changes in the current branch

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — typically a fresh CI-generated pre-release (e.g. `0.8.0-rc.1`) or a stable cut. Reads the auto-generated notes off the GitHub release, classifies and rewrites each…

Review an OpenSpec change bundle for quality, consistency, and implementation readiness. Use when the user wants to analyze specs before implementation, find issues across artifacts, or assess readiness for the apply step. Similar to speckit.analyze but with…

Use when adding a new field to an existing Custom Resource. Guides the full workflow: API types, code generation, CRD examples, client investigation, handler mapping, and tests.

Use when generating testify mocks for an interface. Ensures the interface is registered in .mockery.yml, runs make mocks, and verifies the output file.

Run golangci-lint linters and fix any linting errors. Use when fixing lint issues or before committing code.

Run tests for the edp-keycloak-operator. By default runs unit + integration tests. Use /run-tests e2e for end-to-end tests.

Use when the user asks to enhance, refine, polish, or "look at" the release notes for a tag — typically a fresh CI-generated pre-release (e.g. `0.16.0-rc.1`) or a stable cut. Reads the auto-generated notes off the GitHub release, classifies and rewrites each…

Create a GitHub pull request with Conventional Commits title and structured description. Use when creating PRs, opening pull requests, or submitting code for review.

Create a GitHub issue (Bug Report or Feature Request) for the current project. Use when filing bugs, reporting issues, or requesting features.

Review a GitHub pull request with inline comments on specific diff lines. Use when reviewing PRs, analyzing pull request changes, or providing code review feedback.

Add tests to an Indigo module - core GTest, C API, C++ wrapper, Python, Java, .NET - from templates written in each suite's own style, with the failure contract asserted by type and message and proof that the test catches the regression.

Add or refresh a reference-driven test in Indigo's integration harness (api/tests/integration): choose the comparison, place fixtures, generate references with -u, clean up what the run left, and verify on the second platform and under Jython.

Decide which Indigo test suites a change must pass, know what CI runs for a pull request and what it silently skips, and sort a red run into the change, the build or the machine before calling it a regression.