Agent Skills Hub

list-directory

List files and directories in a specified path to explore project structures, locate files, and verify folder contents.

glyphbox

A framework for an LLM-based NetHack agent that dynamically synthesizes Python code in a secure sandbox to perform complex dungeon exploration and gameplay actions via a high-level API.

Investigating Code Patterns

Systematically trace code flows, locate implementations, diagnose performance issues, and map system architecture to understand complex codebases.

workflow-execute-plans

Execute implementation plans in small, verifiable batches with pause-for-feedback checkpoints to prevent drift and ensure code quality.

syncause-debugger

Evidence-based debugging for Python, Node.js, and Java applications using runtime execution traces and diagnostic MCP tools.

enterprise-agent-builder

Architects enterprise AI agents from structured specs, generating production-ready code, data flow diagrams, and platform-specific logic for ServiceNow, Salesforce, and Snowflake.

familiar-add-tool

Scaffold and register new sensor, actuator, or service tools for familiar-ai, automating file creation and boilerplate integration in agent.py and config.py.

building-ai-agent-on-cloudflare

Build stateful AI agents on Cloudflare Workers using the Agents SDK. Features real-time WebSockets, persistent state management, scheduled background tasks, and native tool integration for production-ready deployments.

writing-skills

A Test-Driven Development (TDD) framework for writing agent skills, using pressure scenarios to ensure documentation guides agent behavior effectively.

sage-mcp-protocol

Sage MCP protocol implementation for integrating external tool servers and standardized AI model context.

atxp

Equip autonomous agents with a funded wallet, identity, and paid API tools for search, generative AI media creation, messaging, and remote communication.

request-optimizer

An intelligent gateway that analyzes, scores, and routes user requests across 27 agents, 27 skills, and 14 MCPs to optimize Claude Code execution.