LLM & Agents

Learn from historical data and build institutional knowledge with Graphiti Memory.Integrates episode storage and retrieval into learning patterns across sessions.Use when user mentions learning workflows, building knowledge over time,analyzing past work patterns, or improving from historical data.

Use after completing a Sea of Nodes chapter implementation, before marking the chapter issue as complete. Validates implementation against official tutorial by comparing node types, test coverage, and key concepts.

Create and maintain AgentV YAML evaluation files for testing AI agent performance. Use this skill when creating new eval files, adding eval cases, or configuring custom evaluators (code validators or LLM judges) for agent testing workflows.

Create or update agent skills following the agent skill specification. Use when user asks to create a new skill or update an existing skill.

Prompt engineering standards and context engineering principles for AI agents based on Anthropic best practices. Covers clarity, structure, progressive discovery, and optimization for signal-to-noise ratio.

Build Optimizely CMS components for Astro v5. Use when creating/modifying components, pages, experiences, or working with GraphQL fragments, opti-type.json, or opti-style.json files (project)

Proactive recovery using plan mode and subagents. After 1-2 failed attempts, STOP trying variations. Enter plan mode and launch parallel Explore/Plan agents to find idiomatic solutions instead of spinning wheels.

To accept or dismiss browser dialogs like alert/confirm/prompt, handle the dialog so automation can continue.

Updates CLAUDE.md based on recent project changes. Use when user says "update docs", "add to CLAUDE.md", "document this", or runs /update-docs command.

Charge la documentation d'un framework depuis son site web dans des fichiers markdown locaux. Supporte Symfony, API Platform, Meilisearch, atournayre-framework et Claude Code.

Claude Code architecture advisor. Classifies knowledge and delegates to appropriate skills. Use when deciding where to put new knowledge or restructuring components.

Agent Skill: Comprehensive Jira integration through lightweight Python scripts. AUTOMATICALLY TRIGGER when user mentions Jira URLs like 'https://jira.*/browse/*', 'https://*.atlassian.net/browse/*', or issue keys like 'PROJ-123'. Use when searching issues (JQL), getting/updating issue details, creating issues, transitioning status, adding comments, logging worklogs, managing sprints and boards, creating issue links, or formatting Jira wiki markup. If authentication fails, offer to configure credentials interactively. Supports both Jira Cloud and Server/Data Center with automatic authentication detection. By Netresearch.

Implement real-time data streaming with Server-Sent Events (SSE), WebSockets, and ReadableStream APIs. Master backpressure handling, reconnection strategies, and LLM streaming for 2025+ real-time applications.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Specialized skill for testing nondominium Holochain applications with Tryorama, providing comprehensive test patterns, agent simulation workflows, performance testing, and multi-agent scenario testing. Use when writing Tryorama tests, debugging zome interactions, or validating Holochain DNA functionality.

Interact with a running Emacs instance via MCP. Use this skill when the user asks to interact with Emacs buffers, set up automation, or coordinate between Claude Code and Emacs. Access Elisp documentation directly via (describe-function 'fn) or (describe-variable 'var).

Create an empty Claude plugin with basic structure and files.

Interactive SQL mode for 30+ databases using DbCli. Provides REPL environment for exploratory queries, rapid prototyping, and database administration. Includes safety prompts before dangerous operations (UPDATE/DELETE/DROP). Use when user wants interactive database session.

Automatically updates terminal window title to reflect the current high-level task. Use at the start of every Claude Code session when the user provides their first prompt, and whenever the user switches to a distinctly new high-level task. Helps developers manage multiple Claude Code terminals by providing clear, at-a-glance identification of what each terminal is working on.

Perform comprehensive code reviews following SOLID principles and best practices