Тег: prompts

Behavior guidelines drawn from Karpathy's observations of how LLMs go wrong when coding: think before acting, keep it simple, make surgical changes only, and drive work by verifiable goals. Can also be used directly as a CLAUDE.md.

Four guidelines based on Karpathy's observations of common LLM coding mistakes: think before coding, prefer simplicity, make only necessary changes, and work toward verifiable goals. Ships as a plugin skill or works as a CLAUDE.md. Community-made, not by Karpathy.

Official Claude Code plugin for the prompts.chat library: search prompts and Agent Skills from your coding assistant, fill in variables and improve prompts with AI. Includes commands, subagents and a remote MCP; saving your own prompts needs an API key.

Makes the agent answer in terse 'caveman' style to cut output tokens, compressing only prose while leaving code, commands and errors intact. Includes a local proxy that compresses logs, test output and other input. Works with Claude Code, Codex and 30+ other agents.

Turns common prompts such as summarizing, extracting insights and rewriting into reusable patterns you pipe text, web pages or YouTube transcripts through, using many model providers or local Ollama. Can also run as a REST API. Run fabric --setup first to configure models.

Plugin for building LLM apps: LangGraph agents with memory and tool use, hybrid-search RAG, vector index tuning, prompt optimization and LLM evaluation. Includes 3 subagents and 8 skills.

Give it a person's name or a vague need and it researches and distills their mental models, decision style and voice into a callable persona skill (e.g. Munger, Feynman, Jobs) to aid analysis and decisions. Works with Claude Code, Codex and other agents.

CLI and library for evaluating and red-teaming LLM apps: compare prompts and model outputs in bulk, scan for security vulnerabilities and plug into CI/CD, with evals running locally. Now part of OpenAI and still open source under MIT.

Distills a colleague, friend or public figure's chats, documents and public material into a persona profile packaged as a callable skill, for handoffs and preserving know-how. Collects material from Lark, DingTalk, Slack and more. Mind privacy and consent when using others' data.

Pressures the agent with big-tech PUA and PIP-style rhetoric plus structured debugging steps and initiative checks, pushing it to change approach after failures and show evidence before claiming done. Modes include always-on, encouragement and P9 lead. Works with Claude Code, Codex and more.

Context and harness engineering skills covering context compression and optimization, multi-agent architectures, memory systems, tool design and LLM-as-judge evaluation, activated automatically by task. Aimed at building production-grade agent systems.

Writes precise prompts for a specific AI tool: identifies the target tool, extracts your intent, asks at most three questions if details are missing, then outputs a lean, paste-ready prompt. Covers dozens of tools including ChatGPT, Claude Code, Cursor, Midjourney and Sora.

Simon Willison's LLM CLI and Python library: call OpenAI, Claude, Gemini and local models from the terminal, with conversations logged to SQLite automatically, plus embeddings, structured extraction, tool calling and a plugin system.

Connects to the Arize Phoenix observability platform so agents can query LLM traces and sessions, manage prompts, browse datasets and read experiment evaluation results. Maintained by Arize; needs a self-hosted or cloud Phoenix instance.

MiniMax's official prompt skill for the Hailuo H3 video model: rewrites video ideas into H3's structured prompt format for text-to-video, first/last frame and full-reference modes, covering visuals, sound effects and music. Docs only; generate videos on Hailuo web, the MiniMax API or a local model.

Turns a theme, a line, an article or photos into a minimal, whitespace-heavy indie-zine portrait poster, and can distill reusable style rules and prompts from reference images, with built-in quality checks. Needs host image generation (e.g. Codex); otherwise outputs prompts only.

Turns a fuzzy UI idea into a design-first prompt spec for UI generation: goal, format, layout, type system, color, imagery, exact copy, constraints and negative prompt, then iterates by changing one variable at a time.

Rewrites a CLAUDE.md so Claude Code follows it better: keeps project identity and map bare at the top, wraps task-specific rules in narrowly conditioned <important if> blocks, cuts snippets and lint-enforceable rules, and keeps every command.

Teaches agents to write video prompts for ByteDance's Seedance 2.0: input limits, @ asset references and camera language, with templates for product ads, short dramas, dance, scenic clips, e-commerce and explainers, in English and Chinese. Prompts only; generate videos with your own account.

Fixes defensive academic writing full of hedging and preemptive disclaimers, organizing abstract, introduction, experiments and conclusion around your strongest contribution. Also helps trim length and self-check before a rebuttal. English and Chinese versions, plus a prompt for any AI chat.

11 skills covering the full AI short-drama and motion-comic pipeline: source analysis, episode scripts, visual design, storyboards, image and video prompts, production, editing and review. Image and video generation needs paid API keys (e.g. Seedance, MiniMax); requires Python 3.9+ and ffmpeg.

Uses unified rule files in a .rulesync directory as the source to generate rules, MCP, commands, subagents, skills and hooks configs for dozens of AI coding tools including Claude Code, Cursor, Copilot and Codex. Can also import existing configs or convert directly between tools.

From product photos and a brief, plans a full e-commerce visual set (main images, Amazon/Shopify detail pages, social promos, livestream scenes) with briefs and image prompts, using 25 scene templates and a consistent style. Renders with an OpenAI-compatible image API key; needs Python 3.10+.

Generates book covers, movie posters, article covers and social images from one sentence, picking colors and composition from 30+ poster masters' styles to suit the subject. Rendering directly needs Python and a PipeLLM image API key; or just generate prompts for another image tool.