Agent Skills Hub

gh-ticket

Create and update comprehensive GitHub issues with full technical context to prevent requirement loss and reduce implementation friction.

long-running-agent

Framework for building AI agents that persist state across multiple context windows, enabling them to complete complex, multi-day coding tasks without losing progress or context.

proofreader-api

Integrates browser-native Proofreader API into web applications for AI-powered text correction, grammar checking, and language support with managed model lifecycle.

saving-codeacts

Preserve successful Python code executions as reusable tools within the gentools package structure, utilizing Pydantic models for structured output and type-safe interfaces.

simplemem-skill

Persistent, semantic long-term memory for AI agents. Save, query, and retrieve cross-session dialogues, decisions, and multimodal context using semantic compression.

moollm

A microworld operating system for LLM-based agent living memory, transforming filesystems into navigable rooms and code into habitable worlds.

litestream

Expert technical support for the Litestream disaster recovery tool, covering WAL monitoring, LTX replication, cloud storage backends, and SQLite page management.

agentdb-state-manager

Persistent state management and workflow analytics using DuckDB for task dependency tracking, historical metrics, and context checkpointing.

parallel-agents

Orchestrate complex workflows by coordinating multiple specialized AI agents for multi-perspective code analysis, feature implementation, and system-wide reviews.

nav-task

Manage Navigator task documentation: create implementation plans, archive completed features, and maintain the task index for Claude Code workflows.

stop

Gracefully pause or close sessions: harvest context, archive work, verify features, and settle AtlasCoin bounties. Use when finishing for the day or wrapping up tasks.

btca-cli

Operate the btca CLI for source-first code research. Manage git, local, and npm resources to ground AI answers in actual codebase context rather than outdated documentation.