dezverev

Starts the AnimalKernel.Web site via run-dev script, uses Context7 for React best practices and Cursor's built-in browser (cursor-ide-browser) for browser-driven development and testing. Use when building or testing the clientapp UI, when the user invokes animal-kernel-web-dev with a follow-up task, or when developing the AnimalKernel.Web frontend with expert UX and browser verification.

Plugin Scaffolding Generator - Creates complete plugin implementations following project patterns. Use when user says "add a plugin for [feature]", "create [Name]Plugin", or "scaffold a new plugin".

Analyzes IntegrationTesterApp test-result JSON (test-results_*.json, multi-model-results_*.json) and produces in-depth analysis of flow (prompt → tools called → response → assertions) and responses, plus actionable fix recommendations. Resolves "latest" or no path to the newest file in src/IntegrationTesterApp/test-results/. Use when the user says /analyze-test-json, "analyze test json", "analyze test results", "analyze latest test results", or @-mentions a test-results or multi-model-results JSON file and asks for analysis or recommendations.

Audits passed tests only: checks that each model response is logically consistent with the user's question (topic, entities, numbers, intent). Does not re-check assertions; adds a semantic-sanity layer so tests that pass on keywords/tools but give off-topic or contradictory answers are surfaced. Resolves JSON like analyze-test-json (test-results_*.json, multi-model-results_*.json in src/IntegrationTesterApp/test-results/). Use when the user says /check-test-logical-resp, "check logical responses", "check test responses make sense", "audit passed tests for logic", or @-mentions a test-result JSON and asks whether responses were logical.

Generates the provider pattern (interface plus fake/real implementations) for plugin data sources. Use when the user says "create a provider for [data source]", "add provider interface for [feature]", or "implement provider pattern for [plugin]".

Picks one workplan from .claude/workplans/Pending/ (skipping plans for issues tagged AIIGNORE or FUTUREWORK), moves it to .claude/workplans/Inprogress/, executes each step via the referenced skills, runs tests and logical-response check. On full success moves to .claude/Done/ and sets the GitHub issue as complete. On partial success (significant progress made, tests pass, but work remains), creates a continuation GitHub issue and progress report. Use when the user says "execute pending workplan", "work from pending", "pick and do a workplan", or "run the next pending plan".

Guides adding a new MCP server integration to AnimalAL-v1: create a nested kernel plugin wrapping the MCP server, add a system message in KernelConstants, register the plugin in StandardKernel, and add integration tests. Use when the user says "add MCP server", "integrate MCP server", "wire MCP tools", or asks how to add or connect a new MCP server to the project.

Discovers open GitHub issues via GitHub MCP (list_issues on user-github) or gh CLI, excludes issues with labels AIIGNORE or FUTUREWORK, prioritizes maint over feat, picks one, analyzes the repo and issue, writes a detailed implementation plan to .claude/workplans/Pending/, commits that plan on the current branch, then sets the issue to "not planned" (closed, state_reason not_planned). Prefer GitHub MCP over gh CLI. Use when the user says "pick an issue and plan", "pick issue and plan", "plan next issue", "choose issue and create plan", or "pick from GitHub issues and plan".

Creates required GitHub labels for issue tracking and workflow automation. Use when setting up a new repository, when labels are missing, or when the user asks to add or sync project labels.

Creates new projects and features following repository structure conventions. Use when the user wants to create a new project, add a new feature, scaffold code, or initialize a new module in this repository.

Executes a pending workplan by priority. Use when the user wants to execute a plan, work on a pending task, start implementation, or continue working on planned issues.

Picks an open GitHub issue by priority and creates an execution plan. Use when the user wants to work on issues, pick a task, plan work, or asks what to work on next.

Writes comprehensive unit tests using xUnit and Moq. Use when the user wants to create tests, add test coverage, write unit tests, or test new functionality. Always finds root cause of failures and fixes the code, not the tests.

Professional React/Node/TypeScript web developer with .NET backend expertise. Uses Context7 for current documentation, verifies all UI work in the browser, and maintains strict visual quality standards. Use when building web features, creating components, implementing APIs, or any frontend/fullstack development task.

Conducts professional code reviews of branch changes and documents findings. Use when the user wants a code review, asks to review changes, check code quality, or before creating a pull request.

Implements feedback from code reviews in the .reviews folder. Use when the user wants to address review comments, fix review findings, implement suggested changes, or respond to code review feedback.

Analyze ColonyGame state, pick next task from design roadmap, implement it end-to-end with tests, commit, and update the development ledger.

Analyze the project and its docs — find missing component docs, update stale ones, ensure every module has accurate documentation in devguide/docs/.

Analyze the current state of ColonyGame from a game design and player experience perspective. Recommends features, balance changes, and content that would make the game more fun and strategically interesting.

Analyze the project and create MCP tools that Claude Code can call to aid development — live game inspection, server control, test running, design tracking.

Performance audit and fix for ColonyGame — profile server tick timing, client rendering FPS, memory usage, and network payload sizes. Identifies bottlenecks and applies fixes.

Commit all pending changes, create a meaningful branch, push, create a PR, and merge it. One-command ship-it workflow.

Report the current state of the ColonyGame project — what's done, what's in progress, what's next, and overall health.

Test coverage auditor for ColonyGame — analyzes feature coverage gaps, writes missing tests, validates edge cases, and ensures all game systems have regression guards.

Full development pipeline — architect a feature, get user approval, then implement, test, and audit it using the mesh agent team.

Develop features for the server meshing system — implement, test against AKS, verify with Playwright, and deploy.

Optimize the server meshing system — benchmark, identify bottlenecks, implement fixes, verify, and deploy against the live AKS cluster.

Ship the current branch — commit, rename, push, create PR, merge to main, and create a new branch for the next piece of work.

Hunt down test coverage gaps in the server meshing system, write tests to fill them, and verify against AKS.

Use when about to do focused reads of unfamiliar implementation files, or when you need a factual answer about files (subsystem map, smallest focused read list, symbol/line anchors, API or flow map, config/docs detail) without pulling large file contents into the main context. Read planning and read-only codebase scouting. Not for code review, bug finding, correctness judgments, edit strategies, or implementation planning.

Get current weather for a location using Open-Meteo