Skill UI

Automatically detects and loads specialized agent instructions from the `AGENTS.md` file in the current working directory. This skill seamlessly integrates project-specific rules and behavioral guidelines into the AI's session context, ensuring the AI operates according to project standards without manual intervention. It is crucial for maintaining specialized agent behaviors across different development tasks and directory changes.

This skill guides the execution of development tasks defined in a structured implementation plan, strictly following the Test-Driven Development (TDD) workflow. It manages the entire development lifecycle, including pre-flight checks, context loading, task identification, writing failing tests, implementing minimal code, refactoring, committing changes, and performing phase-wise verification. Ideal for managing complex, structured software development projects.

A protocol that provides LLM agents with persistent structural memory of a codebase. It maps the project as a directed graph, tracking not only dependencies (what imports what) but also the 'reasons' (why it connects). This allows agents to understand context, perform impact analysis, and navigate large repositories without loading the entire source code into the context window.

Guides Claude Code plugin authors through creating event-driven hooks that validate tool usage, react to outcomes, and gate completion with prompt-based or command hooks, covering PreToolUse, PostToolUse, Stop, Session events, and session context loading.

This comprehensive guide details the end-to-end process of building Retrieval-Augmented Generation (RAG) pipelines using LangChain. It covers critical steps: loading diverse document types (PDF, CSV, TXT), splitting text into manageable chunks, generating high-quality embeddings (OpenAI), storing them in various vector databases (FAISS, Pinecone), and finally constructing the full runnable chain for accurate and context-aware querying.

This pattern addresses the fundamental limitation of context window size by treating the filesystem as a persistent, external memory. Agents can write large outputs, plans, or findings to files, offloading them from the active context. Instead of carrying all information, agents use targeted search and retrieval tools to load only the necessary context on demand, drastically reducing token bloat and improving performance over long, complex tasks.

Copilot Spaces allow users to consolidate and manage project-specific knowledge, including documentation, code snippets, architectural standards, and operational instructions. By loading a Space, Copilot grounds its responses not just on general knowledge, but on the curated, authoritative context of the entire project, enabling accurate Q&A, code generation, and workflow execution based on team standards. Ideal for onboarding, large-scale projects, and maintaining consistency.

Acontext serves as a robust memory layer for production-grade AI agents. It provides persistent sessions, disk-based storage, and sophisticated learning spaces that distill complex conversations into reusable, structured skills. Users can manage the entire system—from project setup and login to skill uploading—via a powerful Command Line Interface (CLI) and API, significantly enhancing agents' contextual understanding and long-term knowledge retention.

This comprehensive tool guides the systematic creation of detailed technical specifications. It orchestrates the entire design lifecycle, including context loading, deep discovery and analysis (tailored for new features, extensions, or simple additions), synthesis using established design principles, and generating a boundary-first, structured design draft. Ensures all requirements are mapped to explicit technical components.

This skill automates the process of transforming high-level product specifications, requirements, and design documents into a detailed, actionable, and executable task plan. It follows a multi-stage workflow, including context loading, parallel research, and two rigorous review gates (Task Plan Review and Task-Graph Sanity Review). The output provides properly sized (1-3 hours), logically sequenced tasks, ensuring every requirement is covered and dependencies are explicitly managed, dramatically improving project planning accuracy.

This skill serves as a comprehensive entry point for initiating new development work. It systematically analyzes project metadata (specs, roadmaps) to determine the optimal work decomposition path (new spec, existing extension, or simple fix). By conducting deep context loading and guiding structured dialogue, it refines vague ideas into clearly defined, actionable scope boundaries. Finally, it proposes multiple concrete technical approaches, complete with trade-offs and estimates, ensuring the resulting project scope is well-defined and ready for implementation.

A tool designed for persistent context management within AI coding sessions. Memory Kit solves the problem of context loss that occurs during compaction or session restarts by saving, loading, updating, and sharing critical session state to a dedicated `MEMORY.md` file. Key functions include creating snapshots of goals and decisions, syncing context across team members via Git, and auditing/pruning stale entries, ensuring continuous workflow and knowledge retention.