agentsclimarketplace
Sign in

Shelves Context & AI engineering

Prompts

1,602 rows from 653 repositories

Writing and improving the instruction itself.

What Prompts skills agree on

563 skill files read, by 542 of the 575 authors on this shelf whose files we hold, 2026-09-06

The middle one of the 78 measured here is ~1.4k tokens long, counted with cl100k_base

Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.

What they tell the agent to do

  1. Provide few-shot examples for complex tasks17 of 542 in 16 filesalso worded as Include few-shot examples for complex tasks; Use few-shot examples for complex output formats
  2. Ask clarifying questions if information is ambiguous16 of 542 in 14 filesalso worded as Ask one clarifying question at a time if information is missing; Ask for clarification if the mode is ambiguous
  3. Output a complete optimized prompt for the user15 of 542 in 9 filesalso worded as output an optimized prompt ready for user execution; output a complete optimized prompt in a code block
  4. Validate structured outputs against schemas15 of 542 in 13 filesalso worded as validate structured output against schema for every run; Use Structured Outputs for validation
  5. Analyze the draft prompt for intent and gaps14 of 542 in 8 files
  6. Detect project tech stack from local files14 of 542 in 8 filesalso worded as detect project tech stack from existing files; detect project tech stack from files
  7. Recommend a model based on task scope13 of 542 in 7 filesalso worded as recommend appropriate models based on task scope; recommend models based on task scope
  8. Present results in the specified output format13 of 542 in 7 filesalso worded as present analysis in the specified output format; Present results in a structured format
  9. Match intent and scope to ECC components13 of 542 in 7 filesalso worded as match intent to ECC commands and skills; map intent and scope to ECC components
  10. Ask one question at a time13 of 542 in 12 filesalso worded as ask one clarifying question at a time; ask one question at a time using AskUserQuestion
  11. Respond in the same language as the user input12 of 542 in 6 filesalso worded as use the same language as the user input; respond in the same language as the user
  12. Ask up to three clarification questions if context is missing11 of 542 in 5 filesalso worded as ask at most three clarifying questions
  13. Use the blueprint skill for epic scope tasks10 of 542 in 4 filesalso worded as use the blueprint skill for epic-scope tasks
  14. Use chain-of-thought for reasoning tasks10 of 542also worded as use chain-of-thought for complex reasoning; use chain of thought for complex reasoning
  15. Advise only and do not execute the task9 of 542 in 3 files

What they tell it not to do

  1. Do not use vague instructions21 of 542also worded as Do not use vague language; Do not use vague instructions like be helpful
  2. Do not write code or create files18 of 542 in 12 filesalso worded as do not write code; do not create files
  3. Do not execute the user task16 of 542 in 10 filesalso worded as do not execute the user's task; do not execute operative tasks
  4. Do not run commands15 of 542 in 9 filesalso worded as do not run bash commands
  5. Do not deploy prompt changes without running evaluations14 of 542 in 8 filesalso worded as do not deploy prompt changes without running evaluations; do not deploy prompt changes without regression testing
  6. Do not trigger on refactoring or performance optimization requests13 of 542 in 7 filesalso worded as do not trigger for performance or code optimization requests; do not trigger on performance or refactoring requests
  7. Do not execute the generated prompt13 of 542also worded as do not execute prompts; do not act on the generated prompt
  8. Do not hardcode prompts in application source code12 of 542 in 8 filesalso worded as do not hardcode prompts in application source code; Do not hardcode the prompt list
  9. Do not trigger if the user says just do it11 of 542 in 5 filesalso worded as do not trigger if the user wants direct execution; do not trigger when user says just do it
  10. Do not change multiple variables simultaneously in variants10 of 542 in 9 filesalso worded as do not change multiple variables simultaneously in variants; Do not make multiple changes simultaneously

What they expect to be installed

  1. git30 of 542
  2. AskUserQuestion26 of 542
  3. grep20 of 542
  4. python317 of 542 in 15 files
  5. python17 of 542 in 14 files
  6. search-first14 of 542 in 8 files
  7. blueprint13 of 542 in 7 files
  8. bash12 of 542
  9. claude12 of 542
  10. node10 of 542

What they ask it to produce

  1. Structured prompt44 of 542 in 43 filesalso worded as XML-structured prompt; structured prompt templates
  2. System prompt36 of 542also worded as System prompts; System prompt instructions
  3. Optimized prompt full version36 of 542 in 31 filesalso worded as optimized prompt quick version; Optimized prompt
  4. Prompt templates19 of 542 in 17 filesalso worded as Final prompt template; Reusable prompt templates
  5. Prompt diagnosis15 of 542 in 9 files
  6. Enhancement rationale14 of 542 in 8 files
  7. Recommended ECC components13 of 542 in 7 files
  8. Few-shot examples13 of 542 in 12 filesalso worded as Few shot examples; Few-shot example set

When Prompts authors say to reach for one

The situations these authors wrote into their own files, counted out of the same 542 authors, with the skills that name each one

How Prompts skills are built

1,465 skill directories by 617 authors, read from their repositories’ own file trees 2026-08-05

The middle bundle among those shipping files is 3 files, 18.1 KB beside SKILL.md

Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.

The shape

  1. SKILL.md is the whole skill225 of 617 authors, 703 of 1,465 skills
  2. files ship beside it392 of 617 authors, 762 of 1,465 skills
  3. executable scripts ship inside126 of 617 authors, 209 of 1,465 skills

The folders they converge on

  1. references/257 of 617 authors, 399 of 1,465 skills
  2. scripts/110 of 617 authors, 184 of 1,465 skills
  3. agents/59 of 617 authors, 71 of 1,465 skills
  4. assets/58 of 617 authors, 77 of 1,465 skills
  5. examples/32 of 617 authors, 42 of 1,465 skills
  6. evals/30 of 617 authors, 72 of 1,465 skills

  • Prompt master

    nidhinjs/prompt-master Skill

    11,108★

    A Claude skill that writes the accurate prompts for any AI tool. Zero tokens or credits wasted. Full context and memory retention

  • Shortfilm prompt

    jnMetaCode/ai-shortfilm-prompts/skills/shortfilm-prompt Skill

    346★ repo

    Claude Code Skill that turns any idea into a cinematic, model-ready video prompt — Sora · Kling · Veo · Seedance. 21 genre templates, 5-stage structure, eval-tested. Distilled from the AI short Hollywood director PJ Ace called "one of the best short films I've seen in years."

  • Prompt architect

    ckelsoe/prompt-architect/skills/prompt-architect Skill

    271★ repo

    Claude Code skill that transforms vague prompts into structured, expert-level prompts using 7 research-backed frameworks (CO-STAR, RISEN, RISE, TIDD-EC, RTF, CoT, CoD)

  • Image

    smixs/visual-skills/image Skill

    165★ repo

    AI film director skills for Claude agents: cinematic dramaturgy (Murch, blocking, montage) + exact prompt syntax for Seedance 2.5, Kling 3.0 Turbo/Omni, Veo 3.1, Nano Banana 2, GPT Image 2. Claude Code, Cursor, Windsurf, OpenCode

  • Vibe creating prompt

    Alisa0808/vibe-creating-skill/skills/vibe-creating-prompt Skill

    no license133★ repo

    Open-source, bilingual AI video-prompt skill — rewrite ideas into model-ready text-to-video prompts. A portable Agent Skill (Claude Code, Codex, OpenClaw, Hermes); generate via Atlas Cloud — Seedance 2.0, Kling, Veo, Hailuo, Wan, Vidu, Gemini Omni, Grok Imagine.

  • Prompt sensei

    chengzhongwei/Prompt-sensei Skill

    107★

    Local-first prompt coach for Claude Code and Codex that improves prompts, observes prompting habits, and analyzes local history with explicit consent.

  • Prompt refine skill

    Apeironics/prompt-refine-skill Skill

    83★

    Agent Skill that silently refines prompts for the currently running model

  • Agent builder

    AlekseiUL/sprut-agent-kit/skills/agent-builder Skill

    64★ repo

    Ready-to-use AI agent with soul, memory, and 23 skills for ClaudeClaw ⚡

  • Skill

    ViryaZheng/promptly-prompt/skill Skill

    48★ repo

    Claude Code skill that forces AI to understand before executing. Three disciplines: cognition check, requirement understanding, method search.

  • Prompt polish

    mfarzanansari/prompt-polish Skill

    42★

    Agent skill that rewrites rough prompts into polished, model-specific prompts — grounded in the official Anthropic prompting guides

  • Agent native architecture

    iliaal/ai-skills/skills/agent-native-architecture Skill

    33★ repo

    Curated collection of agent skills for AI coding assistants.

  • Craft prompt

    V-Songbird/foreman/skills/craft-prompt Skill

    28★ repo

    📋 Prompt-engineering and project-roadmap plugin that crafts professional xml prompts and picks, surveys, and tracks your next tasks.

  • Parable

    SamBWarren/parable/skill/parable Skill

    22★ repo

    A pseudo-Fable mode for Claude Opus — Fable-style planning, verification, and multi-agent discipline as a Claude Code skill set. Unofficial; not affiliated with Anthropic.

  • Prompt warrior

    timoncool/prompt-warrior Skill

    21★

    Agent skill: your AI coding logs as a gamified character sheet — Claude Code, Codex, OpenCode, Gemini, Copilot. 74 achievements, fixed-scale metrics, LLM-written chronicle, research-backed advice. 100% local, RU/EN.

  • Agent system prompt architect

    CR-730/agent-system-prompt-architect-skill/skills/agent-system-prompt-architect Skill

    17★ repo

    Agent Skill for generating deployable system prompts for AI agents

  • Prompt optimization

    shinpr/rashomon/plugins/rashomon/skills/prompt-optimization Skill

    17★ repo

    Measure prompt and skill improvements with blind A/B comparison.

  • Prompt to asset

    MohamedAbdallah-14/prompt-to-asset/io.github.MohamedAbdallah-14/prompt-to-asset MCP server

    17★ repo

    Generate app icons, favicons, OG images, and logos by routing across 30+ image models.

  • Claude design patterns

    codeclawd/fable-mode/skills/claude-design-patterns Skill

    no license16★ repo

    Run Claude Fable 5 on Opus 4.8 in Claude Code. The Mythos-class model pulled by export controls — brought back as a native agentic distillation (FABLE_CODE.md), measured playbook, verification hooks, and design/MCP/test skills. "Fable 5 Lite," done right.

  • Rails llm integration

    nagstler/rails-llm-integration Skill

    13★

    🔥🔥 A Claude Skill that teaches Claude Code how to write LLM features

  • Create gemini agent

    andrader/jup/skills/create-gemini-agent Skill

    no license13★ repo

    jup is a command-line tool for installing and managing agent skills across different ai agents, scopes, and projects.

  • Architecture guardrails

    smixs/disruptor-skills/skills/architecture-guardrails Skill

    13★ repo

    12 agent skills for AI-driven software development — a gated idea-to-ship pipeline: 7w3 design, build-ready specs, executable architecture, honest review loops, a parallel worker-fleet lane, QA, safe deploy. Works with Claude Code, Codex, Cursor and any skills-capable agent. No vibecode.

  • Momentic explore prompt

    momentic-ai/skills/skills/momentic-explore-prompt Skill

    no license13★ repo

    Claude Agent SDK with a E2E testing tool

  • Prompt injection auditor

    screem500/prompt-injection-auditor Skill

    13★

    Security audit skill for LLM agents - prompt injection scanner, attack catalog & defense checklist

  • Ralph prompt generator

    MattMagg/ralph-wiggum-codex/skills/ralph-prompt-generator Skill

    9★ repo

    Installable Codex skill for long-running autonomous coding refinement loops with validation and resumable state

  • Data Pro Skill

    pablodiegoo/Data-Pro-Skill Skill

    7★

    Data Pro Max: The Ultimate AI-Powered Intelligence Suite for Data Science, Survey Analysis, and Professional Reporting.

  • Prompt evaluation runner

    yeaight7/agent-powerups/plugins/agent-evaluation-lab/skills/prompt-evaluation-runner Skill

    6★ repo

    Curated power-ups for coding agents: skills, slash commands, MCP configs, hooks, AGENTS.md templates, and workflows for serious software engineering. Claude Code, Codex, Antigravity CLI, Cursor and more

  • Seedance2 skill

    mightyhuman101/seedance2-skill Skill

    6★

    🎥 Generate precise video prompts for Jimeng Seedance 2.0 using structured patterns, templates, and AI agent compatibility.

  • Optimize claude fable 5 prompts

    IchenDEV/prompt-optimizer-plugins/plugins/claude-fable-5-prompt-optimizer/skills/optimize-claude-fable-5-prompts Skill

    6★ repo

    Dual-protocol prompt optimizer plugins for GPT-5.6 and Claude Fable 5, compatible with Codex and Claude Code.

  • PRISM

    Ronkupper/PRISM Skill

    5★

    A framework for LLM-orchestrated research and audits. Multi-prompt, multi-session, context-disciplined.

  • Token Optimizer Skill

    samibajwaisking/Token-Optimizer-Skill Skill

    no license5★

    Maximum Claude output with minimum tokens. 50+ compression patterns, 10 system prompts, benchmarked results. Star to unlock usage.

  • Prompt shield

    joseph0926/prompt-shield/skills/prompt-shield Skill

    5★ repo

    prompt quality management for Claude Code

  • Prompt improver

    owenob1/prompt-improver/skills/prompt-improver Skill

    5★ repo

    Turn vague agent prompts into precise, verifiable XML specs — then execute them. Headless multi-CLI generator skill (Claude, Grok, Codex, Gemini). Agent Skills + skills.sh compatible.

  • Codex prompt

    TAKEOFF69/codex-skills-kit/skills/codex-prompt Skill

    5★ repo

    Task-typed prompt-engineering Agent Skills for OpenAI Codex.

  • Voice to instrument guide

    stark-ydq/voice-to-instrument-skills/voice-to-instrument-guide Skill

    5★ repo

    Free open-source Agent Skills for voice-to-instrument AI music production — recording best practices, instrument-specific tips, and prompt engineering

  • Ralph prompt generator

    theflixerbox77/ralph-wiggum-codex/skills/ralph-prompt-generator Skill

    5★ repo

    Enable Codex to run objective-first autonomous loops with mandatory review and optional verification for long-running task completion

  • Legnext midjourney

    leggiemint/legnext-skills/legnext-midjourney Skill

    4★ repo

    Lgenext's Midjourney skills

  • Senior saad skill

    a-elradi/senior-saad-skill Skill

    4★

    Elite Claude Code skill for full system architecture & design

  • Ai and agents

    toolbox-playground/super-claudio/skills/ai-and-agents Skill

    4★ repo

    A curated Claude Code plugin with skills for anything Claude can do.

  • Prompt engineering

    hardness1020/awesome-prompt-skill/skills/prompt-engineering Skill

    4★ repo

    Prompt Architecture skills Inspired by Claude Code source code and research catalogs.

  • Review user prompt english

    swapii/claude-linguist/plugins/review-user-prompt-english/skills/review-user-prompt-english Skill

    3★ repo

    Claude Code plugin marketplace with plugins for improving English, Spanish, or any of your other languages

  • Melies cli

    melies-co/melies-cli Skill

    no license3★

    AI Agent Creative Toolkit. 150 AI actors, 220+ visual styles, 50+ AI models. Zero prompt engineering to generate images, videos, thumbnails, movie posters, etc...

  • Ai content filter

    AtulPurohit/Antigravity-Awesome-Skills/plugins/ai-llm-engineering/skills/ai-content-filter Skill

    no license3★ repo

    Installable GitHub library of 300+ professional agentic skills for Claude Code, Antigravity IDE, Gemini CLI, Cursor, and Copilot. Features a custom NPX installer, 9 stack-specific bundles, validation schemas, security auditing, and an interactive catalog explorer app.

  • Epic orchestrator

    chancegraff/rules-for-claude/skills/epic-orchestrator Skill

    no license3★ repo

    Hooks, rules, skills, and instructions for use with Claude Code

  • Prompt fixer

    Soumya7681/prompt-fixer/skills/prompt-fixer Skill

    no license3★ repo

    Cleans up rough prompts before Claude acts on them: fixes typos, expands terse asks into clear instructions, repairs broken {{template}} placeholders, and strips filler openers like "can you / please". Unlike the original CLI, all the actual fixing is done by Claude's own judgment — there's no Node engine to install. The plugin just gives Claude t

  • Image prompt

    veryCoolTimo/imagegen-skills/skills/image-prompt Skill

    3★ repo

    Turn a one-line idea into a production-grade, copy-paste-ready AI image prompt for posters, landing and UI mockups, ads, game art, and logos. A Claude Code skill optimized for gpt-image-2. MIT.

  • Agent creator

    coroboros/agent-skills/skills/agent-creator Skill

    3★ repo

    AI agent skills for Claude Code and compatible agents — workflow, coding, design, Claude Code meta, media, productivity, and writing

  • Agent creator

    anilcancakir/claude-code/plugins/ac/skills/agent-creator Skill

    3★ repo

    Plan-first development partner for Claude Code: interview-driven plans, tier-routed agents (haiku/sonnet/opus), and an adversarial review chain that verifies through real usage.

  • Gpt 5 6 prompt engine

    wp-a/gpt-5.6-prompt-engine/skills/gpt-5-6-prompt-engine Skill

    3★ repo

    Production-grade prompt engineering, migration audits, and eval gates for GPT-5.6 Sol, Terra, and Luna.

  • Prompt forest

    herehere14/Cascade-An-AI-Agent-World-Simulator/extensions/prompt-forest/skills/prompt-forest Skill

    no license3★ repo

    Financial and Economic world prediction simulation. Derived from individual Agent emotions and behaviours using our heart engine, and ripple system can cause different butterfly effects on the simulated world, and make crucial decisions that shapes the macro world, economy, society, company. Achieved over 90% accuracy for covid 19 and 2008 f crises

  • Agent system prompt

    Ahoo-Wang/skills/plugins/ahoo-agent-skills/skills/agent-system-prompt Skill

    3★ repo

    A central aggregation repository for Agent Skills from Ahoo-Wang's open source projects.

  • Empirical prompt tuning

    myksyut/pev-harness/skills/empirical-prompt-tuning Skill

    3★ repo

    Claude Opus 4.7 native Plan-Execute-Verify coding harness — a Claude Code plugin that enforces a 3-phase pipeline with hook-driven verification

  • Prompt keeper

    EA-Studio-SHARK/prompt-keeper/prompt-keeper Skill

    3★ repo

    Capture, auto-classify and reuse your best prompts — a local prompt library inside Claude Code / Cursor / Codex.

  • Prompting buddy

    walkjoi/prompting-buddy Skill

    no license2★

    A Claude Code skill that writes your CLAUDE.md, rules, and agent files — grounded in Anthropic's official best practices. Scan your repo, answer a few questions, get production-ready prompt files.

  • Site to prompt

    dhernz/site-to-prompt/skills/site-to-prompt Skill

    2★ repo

    Convert any site into a prompt — give your agent a URL and it parses the site into a structured prompt another agent can rebuild from.

  • Claude bug bounty

    Mikacr1138/claude-bug-bounty Skill

    2★

    Enable efficient bug bounty hunting across Web2 and Web3 with a tool that supports full recon to detailed reporting.

  • Prompt creator skill

    weiambt/prompt-creator-skill Skill

    no license2★

    一个用于快速生成高质量、结构化提示词的 Claude Code Skill，基于六大提示词模型框架自主分析并动态生成。

  • Prompt engineering

    kongyo2/prompt-engineering-skill/prompt-engineering Skill

    no license2★ repo

  • Agent prompt engineering

    agentailor/skills/agent-prompt-engineering Skill

    2★ repo

    Hand-crafted skills for building AI agents — each one distilled from a technique used and proven in real Agentailor projects.

  • Building deterministic skills

    srinitude/building-deterministic-skills Skill

    2★

    Meta Agent Skill: author deterministic, weak-model-friendly Agent Skills validated against the Hermes runtime. agentskills.io-compliant, with validators and source grounding.

  • En

    ZHOUCOOKIE/prompt-engineering-skill/skill/en Skill

    no license2★ repo

    A Claude Agent Skill that writes, rewrites, reviews, and audits prompts using Anthropic's current official guidance — and picks the model and effort level for you before it drafts. English + 中文.

All 1,602 in the catalog