Skill UI

Specializes in the expert deployment and orchestration of production-grade multi-agent systems using Google's Agent Development Kit (ADK). This skill manages the complex lifecycle, including AgentCard discovery, A2A protocol communication, state management via Memory Bank, and secure execution within the Code Execution Sandbox. Ideal for setting up robust, enterprise-level AI workflows on the Vertex AI Agent Engine, especially during CI/CD pipelines.

Gastown is a robust multi-agent orchestration system designed for Claude Code. It enables the simultaneous execution of multiple AI agents on complex projects. It manages the entire development lifecycle, from setting up workspaces and tracking work items (beads) to managing agent lifecycles (polecats/crew) and merging final code (Refinery). Use it when coordinating complex, parallel AI tasks.

A comprehensive guide to diagnose and resolve common errors in Lindy AI agent workflows. Covers troubleshooting issues related to trigger failures (webhooks, email, schedule), action execution errors (Slack, Gmail, HTTP requests, code), and advanced agent step logic problems, ensuring reliable automation.

This skill thoroughly inspects and validates Vertex AI Agent Engine deployments. It assesses critical components such as the Code Execution Sandbox, Memory Bank, A2A protocol compliance, security posture, and performance metrics. The tool generates weighted production-readiness scores (0-100%) and provides detailed, actionable recommendations to ensure the deployment is robust, secure, and ready for production use.

A workflow that bridges experiment plans to execution. Reads an experiment plan, implements training and evaluation code, performs cross-model code review, deploys runs to GPU clusters, collects initial metrics, and prepares outputs for automated review loops.

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.

This skill provides comprehensive diagnosis and optimization for agent skill libraries. By analyzing historical session transcripts from platforms like Claude Code and Codex, alongside static skill file checks, it scores each skill across 8 dimensions. It identifies issues such as low trigger rates, poor post-invocation user reaction, and incomplete workflow execution, generating a prioritized report (P0/P1/P2) to significantly improve the agent's overall performance and reliability.

This skill provides a detailed, side-by-side comparison of advanced LLM agent evaluation frameworks, such as ruflo and HAL. It helps researchers and developers identify capability gaps—including real code execution, full web browsing, and structured file parsing—and prioritize architectural improvements for robust agent benchmarking and performance tracking.

Delegates coding, code review, and plan review tasks to OpenAI Codex CLI with sandbox controls and security boundaries. Supports autonomous execution, multiple modes, and safe review handling.

Provides comprehensive guidelines for packaging and presenting research artifacts (code, data, models, benchmarks) submitted to NeurIPS. It details requirements for anonymous review packages, public release, and ensures scientific claims are reproducible by providing necessary metadata, licenses, and execution details.

A skill for delegating bounded work to AI agents (Codex CLI, Pi, Claude Code, Hermes) while preserving context, ownership, and progress checks. Covers agent selection, prompt formatting for TUI agents, polling strategies, and remote VPS execution.

This skill provides comprehensive diagnosis and optimization for agent skill libraries. By analyzing historical session transcripts from platforms like Claude Code and Codex, alongside static skill file checks, it scores each skill across 8 dimensions. It identifies issues such as low trigger rates, poor post-invocation user reaction, and incomplete workflow execution, generating a prioritized report (P0/P1/P2) to significantly improve the agent's overall performance and reliability.