Total 50,523 skills, AI & Machine Learning has 8481 skills
Showing 12 of 8481 skills
This skill should be used when the user asks about new features, recent changes, or updates in Claude Code — for example "what's new in Claude Code?", "Claude Code changelog", "what did I miss in Claude?", "any recent updates?", "tell me about new Claude features", or "what's changed since version 1.0.30?". It fetches the official changelog, filters for notable features (excluding bug fixes), researches each feature for deeper context on Anthropic's website, and presents mini-article summaries. Supports both automatic tracking (since last check) and explicit version queries.
Comprehensive documentation audit and generation. Launches parallel agents for high-level docs, module-level docs, decision records, and state diagrams. Use when: documentation gaps, post-implementation docs, README updates, architecture docs.
Detect backtest iteration stagnation and generate structurally different strategy pivot proposals when parameter tuning reaches a local optimum.
Comprehensive guide for skill development based on Anthropic's official best practices - use for complex skills requiring detailed structure
NEVER escalate without investigation first. This is the Iron Law. Use when evaluating whether to escalate models, facing genuine complexity requiring deeper reasoning, novel patterns with no existing solutions, high-stakes decisions requiring capability investment. Do not use when thrashing without investigation - investigate root cause first. DO NOT use when: time pressure alone - urgency doesn't change task complexity. DO NOT use when: "just to be safe" - assess actual complexity instead.
Claude Code hooks configuration and development. Covers hook lifecycle events, configuration patterns, input/output schemas, and common automation use cases. Use when user mentions hooks, automation, PreToolUse, PostToolUse, SessionStart, SubagentStart, PermissionRequest, WorktreeCreate, WorktreeRemove, TeammateIdle, TaskCompleted, ConfigChange, or needs to enforce consistent behavior in Claude Code workflows.
Used when you need to perform Discover (reverse engineering) on legacy projects with existing code, consolidate repository facts into `.aisdlc/project/`, and you find that AI or teams frequently guess entry points and boundaries, have duplicate writing of indexes and details, or lack evidence chains leading to repeated rework.
Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications with detailed traces, running evaluations on datasets, or monitoring production AI systems with real-time insights.
Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.
QQ Bot Media Sending Guide. Teaches AI how to send images to users. [Important] The <qqimg> tag must be used when users request to send images.
Phase-level planning workflow for planner agents. Handles reading templates, exploring codebase references, creating plan.md and phase files, self-validation, and checkpoint reporting to the orchestrator. Invoke this skill as your first action — not user-invocable.
Migrate memory blocks from an existing agent to the current agent. Use when the user wants to copy or share memory from another agent, or during /init when setting up a new agent that should inherit memory from an existing one.