Agent Skills Hub

agent-canvas

Orchestrate visual communication by drawing diagrams, flowcharts, and annotations on a TLDraw canvas via CLI. Ideal for architectural planning, PR reviews, and logging agent output.

recursive-decomposition

Handles large-scale tasks by automatically breaking them down into manageable, recursive sub-tasks to overcome context window limits and improve reasoning accuracy on large codebases and document sets.

linear

Linear issue management and synchronization for LobeHub, featuring automated PR referencing, sub-issue tree decomposition, and status tracking.

development

Comprehensive Linux development environment management for compilers, build tools, IDEs, and debugging workflows.

docstring

Generate and update PyTorch-compliant function and method docstrings using reStructuredText/Sphinx conventions.

Global Coding Style

Apply reality-first coding standards: intentional naming, focused functions, guard clauses, and deterministic side effects, with no speculative features.

handoff

Create structured session handoff documents to ensure continuity when switching contexts, taking breaks, or ending your workday.

chrome-devtools-mcp

Chrome DevTools MCP server for AI-driven browser automation, testing, and debugging via Puppeteer. Features input automation, visual snapshots, performance tracing, and network inspection.

interface-design

Interface design guidance for utilitarian apps, focusing on dashboards, admin panels, and data-heavy UIs using a component-library-first approach.

requirement

Write high-quality user stories and requirement documents following the INVEST criteria.

swift-review

Expert Swift code review for macOS/iOS. Detects memory leaks, threading bugs, concurrency issues, and accessibility gaps using parallel analysis agents.

tldr

Token-efficient codebase analysis skill for call graphs, semantic search, impact analysis, and data flow. Saves ~95% tokens vs. raw reads.