Skill UI

Generates a precise, evidence-based installation plan for a project's component catalog (ECC). Instead of loading a full, noisy bundle, this skill analyzes the repository's actual code structure, dependencies, and CI/CD evidence to classify skills, rules, and commands into 'DAILY' (essential) and 'LIBRARY' (referenceable) buckets. It ensures only necessary components are active, optimizing the development environment.

This guide addresses the critical difference between standard Node.js SHA3-256 and the Keccak-256 hash function required by Ethereum. Using the wrong hash function can silently corrupt calculated function selectors, event topics, storage slots, and derived addresses. It provides best practices and code examples for using dedicated libraries (like ethers, viem, web3.js) to ensure correct, Ethereum-compliant hashing in JavaScript and TypeScript environments.

A comprehensive guide to structuring Flutter applications using a feature-first, layered architecture. It defines clear separation of concerns across UI, Logic (Domain), and Data layers, ensuring maintainability and testability. This approach guides developers on organizing code by feature modules, making it ideal for building complex, scalable, and long-lived applications while remaining agnostic to specific state management solutions.

This utility compresses verbose natural language notes, memories, and documents (like markdown files) into a highly dense format. By eliminating filler words, redundant phrases, and grammatical fluff, it significantly reduces the token count for large inputs. Crucially, it strictly preserves all technical content, including code blocks, inline code, URLs, file paths, and commands, ensuring technical accuracy while boosting AI processing efficiency.

This skill provides comprehensive interaction with the DeerFlow AI agent platform via its robust HTTP API. It is designed for managing complex, multi-step AI workflows, including deep academic research, code execution, web browsing, and file analysis. Users can initiate and continue conversation threads, manage agent memory, upload documents (PDF, etc.), and delegate complex research tasks to the autonomous agent system.

A comprehensive skill for engineering high-quality, observable Go services. It guides best practices across five key signals: structured logging (slog), Prometheus metrics, OpenTelemetry tracing, continuous profiling (pprof), and RUM. Use this when instrumenting new features, reviewing PRs, or auditing entire codebases to ensure systems are fully measurable for reliable diagnosis and alerting.

A comprehensive guide for structuring, organizing, and initializing Go projects. It covers architectural decisions (Clean, Hexagonal, DDD), best practices for directory layouts (CLI, Service, Library, Monorepo), mandatory naming conventions, and adhering to 12-Factor App principles. Use this when starting a new Go service, refactoring large codebases, or setting up a complex monorepo.

samber/oops provides a structured, context-rich replacement for standard Go error handling. It allows engineers to attach critical metadata—such as domain, user ID, stack traces, error codes, and request details—directly to the error object. This ensures that variable data travels with the error through the call stack, enabling advanced APM tools (like Sentry or Datadog) to group and diagnose issues accurately, treating errors as structured, actionable data.

A comprehensive skill for interacting with Lark Whiteboard documents within FeiShu. It allows users to query whiteboard content, export as images or structured code (Mermaid/PlantUML), edit raw nodes, and update content using various formats like OpenAPI, Mermaid, or PlantUML. Ideal for visualizing architecture, processes, or organizational structures.

This mode simulates a P9 Tech Lead role, specializing in strategic project management and complex task orchestration. It does not write code itself, but focuses on generating detailed, structured Task Prompts (six elements) and managing the entire workflow delivery from multiple specialized agents (P8 teams). Ideal for coordinating large, complex projects requiring strategic decomposition and coordination of three or more parallel agents.

This tool generates ultra-compressed, standardized commit messages adhering strictly to the Conventional Commits specification. It analyzes code changes (diffs) to extract the core intent, structuring the output into a precise subject line and an optional body. It ensures that commit history remains clean, highly readable, and instantly understandable, significantly reducing noise and maintaining professional version control standards.

Streamlines the process of providing code review feedback by enforcing a highly structured, ultra-compressed format. It guides users to cut through noise and deliver actionable, single-line comments (Location: Problem. Fix.) while maintaining critical context like severity and technical rationale. Ideal for team settings prioritizing efficiency and high code quality.