Loading...
Loading...
Found 378 Skills
Generates prompts for storyboard sketches. Invoked when users need to create prompts for storyboard sketches for film, animation or video projects.
Targeted Chat Room: Recommend experts based on topics or accept user-specified experts to simulate multi-role conversations. Trigger methods: /定向聊天室, 「定向聊天室」
Generate and edit high-quality images with AI. Emphasize strong prompt design, structured JSON prompting, reference-image workflows, text rendering, and iterative refinement. Use any time the user needs an image generated.
Mandatory protocol for dispatching any built-in and custom agent in this project via the task tool. Use this skill EVERY TIME you are about to call the task tool with a custom agent_type. This skill ensures the agent's intended model (declared in its YAML frontmatter) is respected rather than overridden by a default. Also encodes prompting best practices for subagent context and quality. ALWAYS invoke before any task tool call that targets a custom agent — even if the agent name seems obvious.
Use when starting any conversation, receiving a new task, or when uncertain which skill applies - establishes how to find and use all 64 toolkit skills, requiring Skill tool invocation before ANY response including clarifying questions
Review a prompt to identify ambiguities, missing constraints, and hallucination risks, and provide an optimized version.
Design Pydantic models and LLM prompt templates for structured extraction pipelines. Use when creating, editing, or reviewing Pydantic models that serve as LLM output schemas, or when writing prompt templates that pair with those models. Trigger: "pydantic model", "structured output", "extraction schema", "LLM output model", "schema design".
Autonomously optimize an existing AI skill by running it repeatedly against binary evals, mutating one instruction at a time, and keeping only changes that improve pass rate. Based on Karpathy-style autoresearch, but applied to SKILL.md iteration instead of ML training. Use when optimizing a skill, benchmarking prompt quality, building evals for a skill, or running self-improvement loops on reusable agent instructions. Triggers on: skill-autoresearch, optimize this skill, improve this skill, benchmark this skill, eval my skill, run autoresearch on this skill, self-improve skill.
A prompt repetition technique for improving LLM accuracy. Achieves significant performance gains in 67% (47/70) of 70 benchmarks. Automatically applied on lightweight models (haiku, flash, mini).
Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.
Transform code, issues, or context into a detailed prompt/context for another LLM to fix or implement. Use when preparing comprehensive context for external LLM assistance, bug fixes, improvements, or feature implementations. Provides detailed context without implementation suggestions, letting the receiving LLM decide how to implement solutions. Focuses on "what" (problem, requirements, current state) not "how" (implementation approach).
Ultra-compressed communication mode. Distills AI output to its core — ~65% fewer tokens while keeping full technical accuracy. Use when user says "caveman", "terse", "kurz", "kurz&knapp", "kurz bitte", "less tokens", "weniger text", or invokes /caveman-distillate.