Marketplace des Skills d’Agent

指导从 Entity 到 Controller 的完整 CRUD 功能开发流程。包括实体创建、DTO设计、服务实现、控制器开发、数据库配置和迁移。当用户需要开发新的 CRUD 功能、创建 API 接口、或添加新的业务模块时使用。

Delegate coding tasks to Google Jules AI agent for asynchronous execution. Use when user says: 'have Jules fix', 'delegate to Jules', 'send to Jules', 'ask Jules to', 'check Jules sessions', 'pull Jules results', 'jules add tests', 'jules add docs', 'jules review pr'. Handles: bug fixes, documentation, features, tests, refactoring, code reviews. Works with GitHub repos, creates PRs.

Brainstorm and plan features through codebase search, technology research, and 2-3 approach comparison before producing implementation-ready plans. Use when starting features, designing solutions, or planning complex work. Triggers on: whiteboard, let's plan, brainstorm, design this, figure out how to build. Save plans to docs/plans/ for execution via /code-foundations:building.

Generate and transcribe speech using Google's Gemini-TTS and Chirp 3 models. Supports Text-to-Speech (Single/Multi-speaker), Instant Custom Voice, and Speech-to-Text (Transcription/Diarization).

Modify CDK Step Functions workflows (ReviewProcessor and ChecklistProcessor) for the RAPID application, including workflow step changes, Map State concurrency, retry/timeout configuration, error handling, and parameter updates. Use when changing workflow step sequences, adjusting concurrency, modifying retry logic, or adding/removing workflow steps.

Replace complex conditional logic with table lookups. CHECKER mode identifies opportunities where if/else chains or inheritance hierarchies should become tables. APPLIER mode designs table structure, access method (direct/indexed/stair-step), and key transformations. Use when writing 4+ if/else branches, switch statements keep growing, subclasses differ only in data not behavior, or data changes without code changes. Triggers on: too many if statements, switch growing, subclass per type, customer-controlled format, data-driven behavior, lookup table, mapping data to actions.

This skill should be used when the user asks to "scaffold a Bankr project", "create new Bankr bot", "build a Bankr web service", "create Bankr dashboard", "build Bankr CLI tool", "project structure for Bankr", "Bankr project types", or needs guidance on directory structures and templates for different types of Bankr API integrations.

Create a long-term memory in Elroy

Guide systematic debugging using scientific method: STABILIZE -> HYPOTHESIZE -> EXPERIMENT -> FIX -> TEST -> SEARCH. Two modes: CHECKER audits debugging approach (outputs status table with violations/warnings), APPLIER guides when stuck (outputs stabilization strategy, hypothesis formation, fix verification). Use when encountering ANY bug, error, test failure, crash, wrong output, flaky behavior, race condition, regression, timeout, hang, or code behavior differing from intent. Triggers on: debug, fix, broken, failing, investigate, figure out why, not working, it doesn't work, something's wrong.

Generate, edit, and compose images using Gemini Nano Banana models via portable Python scripts. Handles authentication via API Key or Vertex AI environment variables. Available parameters: prompt, model, aspect-ratio, safety-filter-level. Always confirm parameters with the user or explicitly state defaults before running.

Plutonium Rodauth integration - authentication setup, account types, and configuration

Plutonium resource controllers - CRUD actions, customization, and integration

Configure tree-sitter CLI and language grammars for AST-powered code review. Use when AST extraction fails, tree-sitter not found, grammars missing, or setting up new machine. Triggers on: setup tree-sitter, install grammars, AST not working, tree-sitter not found, setup ast.

Enforce Design-It-Twice workflow: generate 2-3 radically different approaches, compare them, then implement. Use when designing modules, APIs, or classes before implementation. Triggers on: design, create class, add module, implement feature, new service, API design, before implementing. Produces structured design document with approaches, comparison table, choice rationale, and depth check.

Use when working in projects that use state-ref; follow proxy-based reactivity and lens pattern guidelines.

Use this skill when creating Model Context Protocol (MCP) servers, adding tools to extend Claude's capabilities, or integrating external APIs as MCP tools.

Ingest documents into Elroy memory

Interact with Google Chat - list spaces, send messages, read conversations, and manage DMs. Use when user asks to: send a message on Google Chat, read chat messages, list chat spaces, find a chat room, send a DM, or create a new chat space. Lightweight alternative to full Google Workspace MCP server with standalone OAuth authentication.

Modify the review-item-processor agent's prompts, model configuration, confidence thresholds, JSON output schemas, and tool setup. Use when changing document or image review prompts, adjusting model IDs, modifying output format, updating citation settings, or adding new tools to the agent.

Optimize performance through mandatory measure-first workflow with 3 stages: MEASURE (gate) -> FUNDAMENTAL FIX -> CRITICAL PATH REDESIGN. Use when code is too slow, laggy, hangs, doesn't scale, has timeouts, OOM errors, or high CPU/memory. Emphasize that simpler code usually runs faster. Triggers on: performance issues, optimization needed, takes forever, profiling. Produces profiling-backed optimization with verified improvement.

Persistent memory system for AI agents. Use this skill to remember context across conversations, recall relevant information, and build long-term knowledge. Activate when you need to store decisions, learnings, errors, or context that should persist beyond the current session.