Uses a structured two-phase process (analysis then implementation) to build features according to project conventions.
Install
mkdir -p .claude/skills/create && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/19157" && unzip -o skill.zip -d .claude/skills/create && rm skill.zipInstalls to .claude/skills/create
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Plan, implement, and test a new feature or change with pre-implementation reviewKey capabilities
- →Review tasks against project best practices and Angular-way conventions
- →Identify existing code for reuse and shared logic for extraction
- →Ask clarifying questions for ambiguous requirements
- →Present alternative implementation approaches with trade-offs
- →Verify plans address task requirements and project conventions
- →Implement features, run lint and build, and report completion
How it works
The skill performs an analysis phase to review tasks, check for code reuse, and clarify requirements before implementation. It then proceeds with feature implementation and testing after user approval.
Inputs & outputs
When to use create
- →Plan a new feature
- →Review feature requirements
- →Implement a bug fix
About this skill
Create
Plan, implement, and test a feature or change following project conventions.
Task
$ARGUMENTS
If $ARGUMENTS is empty, use the task described in the preceding conversation messages.
Workflow
Phase 1 — Analysis
Switch to plan mode and perform the following analysis before writing any code.
1.1 Task review
- Does the task align with Angular-way and project best practices (see CLAUDE.md)?
- Are there simpler alternatives — easier to implement, maintain, or extend, or more popular in the Angular ecosystem?
- If alternatives exist, present them to the user with trade-offs and let them choose.
1.2 Codebase reuse check
- Search the project for existing code that can be reused (services, components, utilities, patterns).
- Identify shared logic that should be extracted to avoid duplication across components.
1.3 Clarifications
- Ask clarifying questions at any point if requirements are ambiguous.
- When multiple implementation approaches exist, present options and let the user decide — never choose for the user.
1.4 Plan self-review
Before presenting the plan to the user, verify:
- Does it fully address the task requirements?
- Does it comply with all project conventions (CLAUDE.md)?
- Is the plan sufficient and complete — no missing steps?
- Is it not over-engineered — no unnecessary abstractions or premature generalization?
If issues are found, fix them or ask additional questions before presenting.
Phase 2 — Implementation
After the user approves the plan:
- Implement the feature following the approved plan
- Run
yarn lintandyarn build— fix any issues found - Verify the implementation complies with project conventions (CLAUDE.md)
- Report completion to the user
Phase 3 — Testing
Wait for the user to confirm that the feature works correctly before writing tests.
After confirmation:
- Write unit tests (Jest + Spectator) covering the new functionality
- Run
yarn lintandyarn nx test <project>— fix any issues found - Report test results
Rules
- All conversation in Russian
- Follow all conventions from CLAUDE.md strictly
- Never skip the analysis phase — even for seemingly simple tasks
- Never write tests before user confirms the implementation works
When not to use it
- →When the analysis phase is to be skipped
- →When tests are to be written before user confirmation
- →When project conventions from CLAUDE.md are not followed
Limitations
- →All conversation must be in Russian
- →The analysis phase cannot be skipped
- →Tests cannot be written before user confirms implementation works
How it compares
This workflow integrates a formal analysis and planning phase, including self-review and user consultation, before any code is written, unlike a direct coding approach.
Compared to similar skills
create side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| create (this skill) | 0 | 5mo | No flags | Intermediate |
| codex | 32 | 2mo | Review | Advanced |
| senior-fullstack | 35 | 7mo | Review | Intermediate |
| typescript-write | 30 | 1mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
codex
Lucklyric
Invoke Codex CLI for complex coding tasks requiring high reasoning capabilities. This skill should be invoked when users explicitly mention "Codex", request complex implementation challenges, advanced reasoning, or need high-reasoning model assistance. Automatically triggers on codex-related requests and supports session continuation for iterative development.
senior-fullstack
davila7
Comprehensive fullstack development skill for building complete web applications with React, Next.js, Node.js, GraphQL, and PostgreSQL. Includes project scaffolding, code quality analysis, architecture patterns, and complete tech stack guidance. Use when building new projects, analyzing code quality, implementing design patterns, or setting up development workflows.
typescript-write
metabase
Write TypeScript and JavaScript code following Metabase coding standards and best practices. Use when developing or refactoring TypeScript/JavaScript code.
codex-skill
feiskyer
Use when user asks to leverage codex, gpt-5, or gpt-5.1 to implement something (usually implement a plan or feature designed by Claude). Provides non-interactive automation mode for hands-off task execution without approval prompts.
subagent-driven-development
davila7
Use when executing implementation plans with independent tasks in the current session
at-dispatch-v2
pytorch
Convert PyTorch AT_DISPATCH macros to AT_DISPATCH_V2 format in ATen C++ code. Use when porting AT_DISPATCH_ALL_TYPES_AND*, AT_DISPATCH_FLOATING_TYPES*, or other dispatch macros to the new v2 API. For ATen kernel files, CUDA kernels, and native operator implementations.