# Loreto

> Loreto turns your domain expertise into Claude Code skills. Generate a skill from a YouTube video, article, PDF, image, or your own workflow; run it as a CLI + MCP; or buy and sell skills in the marketplace. Skills are discoverable and installable by humans and AI agents.

Loreto is the marketplace and runtime for Claude Code skills. Every skill is a packaged SKILL.md with references and a runnable test that drops into .claude/skills/. Skills can be installed via MCP, run as a CLI, and bought or sold — agents pay in USDC via x402.

## Start here
- [Generate a skill](https://loreto.io/): Turn any source into a Claude Code skill.
- [Marketplace](https://loreto.io/marketplace): Browse and buy Claude Code skills.
- [Free skills](https://loreto.io/free-skills): 22 skills you can download free.
- [Pricing](https://loreto.io/pricing): Plans and pay-per-call via x402.
- [Docs](https://loreto.io/docs): API, MCP, x402, and skill generation.

## Free tools
- [SKILL.md linter & generator](https://loreto.io/tools/skill-md-linter): Write or validate a Claude Code SKILL.md — correct frontmatter, structure, and naming, checked live.
- [Claude token & cost calculator](https://loreto.io/tools/claude-cost-calculator): Estimate tokens and USD cost of a Claude API call across every current model.
- [llms.txt generator](https://loreto.io/tools/llms-txt-generator): Generate a spec-compliant llms.txt so GPTBot, ClaudeBot and Perplexity understand your site.

## Free skills
- [Fine-Tuning GPT Models](https://loreto.io/skills/fine-tuning-gpt-models)
- [Preprocessing and Tokenizing Textual Datasets](https://loreto.io/skills/preprocessing-and-tokenizing-textual-datasets)
- [Training GPT Models with PyTorch](https://loreto.io/skills/training-gpt-models-with-pytorch)
- [harness-shape: judge an agent framework by its imports](https://loreto.io/skills/harness-shape-judge-an-agent-framework-by-its-imports)
- [HyDE, Federated and Streaming RAG](https://loreto.io/skills/hyde-federated-and-streaming-rag)
- [Agent Driven Code Editing](https://loreto.io/skills/agent-driven-code-editing)
- [Agent Context Files](https://loreto.io/skills/agent-context-files)
- [Skunky: Local-First Compartmentalized LLM Work](https://loreto.io/skills/skunky-local-first-compartmentalized-llm-work)
- [Branded ECharts Visualizations](https://loreto.io/skills/branded-echarts-visualizations)
- [Optimizing for People Also Ask](https://loreto.io/skills/optimizing-for-people-also-ask)
- [Fixing JavaScript Rendering for Crawlers](https://loreto.io/skills/fixing-javascript-rendering-for-crawlers)
- [Consolidating Cannibalizing Content](https://loreto.io/skills/consolidating-cannibalizing-content)
- [Designing a Programmatic SEO Template](https://loreto.io/skills/designing-a-programmatic-seo-template)
- [Tracking Citation Position & Prominence in AI Answers](https://loreto.io/skills/tracking-citation-position-prominence-in-ai-answers)
- [Implementing Article & Author (E-E-A-T) Schema](https://loreto.io/skills/implementing-article-author-e-e-a-t-schema)
- [Detecting AI Bot Traffic in Server Logs](https://loreto.io/skills/detecting-ai-bot-traffic-in-server-logs)
- [Scoring Content Citation-Worthiness for AI Answers](https://loreto.io/skills/scoring-content-citation-worthiness-for-ai-answers)
- [Authoring an llms.txt for Agent Routing](https://loreto.io/skills/authoring-an-llms-txt-for-agent-routing)
- [Agentic RAG Routing](https://loreto.io/skills/agentic-rag-routing)
- [Temporal Reasoning Sleuth](https://loreto.io/skills/temporal-reasoning-sleuth)
- [Routing Work Across AI Harnesses](https://loreto.io/skills/routing-work-across-ai-harnesses)
- [Detecting Harness Lock-in](https://loreto.io/skills/detecting-harness-lock-in)

## Popular skills
- [Agentic RAG Fallback](https://loreto.io/skills/agentic-rag-fallback): Designs a Retrieve-Check-Route fallback pattern for agentic (multi-tool) RAG systems so that when every retri…
- [Agentic RAG Routing](https://loreto.io/skills/agentic-rag-routing): Designs an agentic RAG architecture where an LLM agent acts as a dynamic router, semantically selecting the c…
- [Auditing Intelligence-Context Fit](https://loreto.io/skills/auditing-intelligence-context-fit): Audits the fit between a model's reasoning capability and the complexity of the context it receives. Use when…
- [Benchmarking AI Agents Beyond Models](https://loreto.io/skills/benchmarking-ai-agents-beyond-models): Separates harness contribution from model contribution when evaluating AI agent performance, so that benchmar…
- [Designing Hybrid Context Layers](https://loreto.io/skills/designing-hybrid-context-layers): Designs hybrid AI context architectures that combine RAG, knowledge graphs, episodic memory, and long-context…
- [Detecting Harness Lockin](https://loreto.io/skills/detecting-harness-lockin): Identifies and quantifies the compounding switching cost of an AI coding agent harness commitment before it b…
- [Diagnosing RAG Failure Modes](https://loreto.io/skills/diagnosing-rag-failure-modes): Diagnoses RAG system failures by classifying queries as factual-lookup-safe vs. relational-temporal (where RA…
- [Evaluating AI Harness Dimensions](https://loreto.io/skills/evaluating-ai-harness-dimensions): Evaluates AI coding agent platforms across five structural dimensions — execution philosophy, state and memor…
- [RAG Pipeline Vector DB](https://loreto.io/skills/rag-pipeline-vector-db): Designs a traditional RAG pipeline organized around the Query-Retrieve-Augment-Generate (QRAG) flow and its s…
- [RAG Pipeline Vector Store](https://loreto.io/skills/rag-pipeline-vector-store): Designs a baseline single-hop RAG pipeline that grounds LLM answers in private or domain-specific documents v…
- [RAG Query Routing](https://loreto.io/skills/rag-query-routing): Designs and implements LLM-backed query routing across multiple domain-specific vector stores using the retri…
- [RAG Relevance Fallback](https://loreto.io/skills/rag-relevance-fallback): Designs a conditional fallback branch in a single-store RAG pipeline so that when retrieved documents are irr…
- [Routing Work Across AI Harnesses](https://loreto.io/skills/routing-work-across-ai-harnesses): Designs hybrid agent workflows that route tasks intelligently between AI coding agent harnesses based on task…
- [Synthesizing Institutional Knowledge](https://loreto.io/skills/synthesizing-institutional-knowledge): Builds organizational memory systems that capture decision provenance, causal chains, and institutional conte…
- [Temporal Reasoning Sleuth](https://loreto.io/skills/temporal-reasoning-sleuth): Engineers temporal reasoning capabilities for AI agents — enabling them to trace decision chains, reconstruct…

## Comparisons
- [Claude Code skills vs MCP servers: what's the difference?](https://loreto.io/compare/claude-code-skills-vs-mcp-servers)
- [Loreto vs writing Claude Code skills by hand](https://loreto.io/compare/loreto-vs-building-skills-by-hand)
- [RAG vs knowledge graph: which does your agent need?](https://loreto.io/compare/rag-vs-knowledge-graph)
- [Skill package vs CLI + MCP: which artifact should Loreto build?](https://loreto.io/compare/skill-package-vs-cli-mcp)
- [Loreto vs GitHub gists for sharing Claude Code skills](https://loreto.io/compare/loreto-vs-github-gists-for-skills)
- [Claude Code skills vs custom instructions / CLAUDE.md](https://loreto.io/compare/claude-code-skills-vs-custom-instructions)
- [SEO vs AEO: optimizing for Google vs AI answer engines](https://loreto.io/compare/seo-vs-aeo)
- [Generate a skill vs buy one from the marketplace](https://loreto.io/compare/generate-vs-buy-a-skill)
- [Loreto skills vs a prompt library](https://loreto.io/compare/loreto-vs-prompt-library)

## By use case
- [Claude Code skills for RAG engineers](https://loreto.io/for/rag-engineers)
- [Claude Code skills for AI platform teams](https://loreto.io/for/ai-platform-teams)
- [Claude Code skills for SRE & incident response](https://loreto.io/for/sre-teams)
- [Claude Code skills for SEO & AEO teams](https://loreto.io/for/seo-teams)
- [Claude Code skills for solo developers & indie hackers](https://loreto.io/for/solo-developers)
- [Claude Code skills for content teams](https://loreto.io/for/content-teams)
- [Claude Code skills for growth & SEO marketers](https://loreto.io/for/growth-marketers)
- [Claude Code skills for ML & evaluation engineers](https://loreto.io/for/ml-engineers)
- [Claude Code skills for AI agent builders](https://loreto.io/for/agent-builders)
- [Claude Code skills for agencies & consultants](https://loreto.io/for/agencies)

## Guides
- [Cursor, Claude Code, Cline — why does the same model feel completely different in each?](https://loreto.io/blog/harness-vs-model-cursor-claude-cline)
- [Your model got smarter and your answers got worse. Why?](https://loreto.io/blog/model-got-smarter-answers-got-worse)
- [Why can't your AI agent answer 'how did we get here?' even when it has all the data?](https://loreto.io/blog/why-ai-agents-fail-temporal-reasoning)
- [Why does every new hire ask the same questions even when you have the docs?](https://loreto.io/blog/why-new-hires-ask-same-questions)
- [When should you stop adding chunks to your RAG and reach for a knowledge graph?](https://loreto.io/blog/stop-adding-chunks-knowledge-graph)
- [Why does RAG keep failing on questions about your team's history?](https://loreto.io/blog/why-rag-fails-on-team-history)
- [When you have several vector stores, how do you send a query to the right one?](https://loreto.io/blog/route-query-to-right-vector-store)
- [What's the smallest RAG pipeline that actually works — and why is it exactly one LLM call?](https://loreto.io/blog/smallest-rag-pipeline-that-works)
- [Your RAG answers are wrong. Is it the model — or your chunking?](https://loreto.io/blog/rag-answers-wrong-model-or-chunking)
- [What should your agent do when every retriever tool comes back empty?](https://loreto.io/blog/agent-every-retriever-empty)
- [When your RAG finds nothing relevant, should it stay silent or fall back to the model?](https://loreto.io/blog/rag-found-nothing-relevant-fallback)
- [Should the agent decide what to retrieve, or should the pipeline?](https://loreto.io/blog/agent-decides-what-to-retrieve)
- [Should one agent do the whole job, or should you split it?](https://loreto.io/blog/one-agent-or-several-for-the-job)
- [How locked in are you to your AI coding tool?](https://loreto.io/blog/how-locked-in-are-you-to-your-ai-tool)
- [Was that the model, or was it the harness?](https://loreto.io/blog/is-the-model-good-or-is-the-harness-good)
- [Why does your agent forget everything between sessions?](https://loreto.io/blog/agent-forgets-everything-between-sessions)
- [Is this a prompt, or is it a skill?](https://loreto.io/blog/is-this-a-prompt-or-a-skill)
- [Why does your agent say it's done when it isn't?](https://loreto.io/blog/agent-says-done-but-it-isnt)
- [Which model should this task actually use?](https://loreto.io/blog/which-ai-model-should-this-task-use)
- [Why does your agent keep following a rule you retired?](https://loreto.io/blog/why-your-agent-follows-an-outdated-rule)
- [What should an agent read at the start of a six-hour run?](https://loreto.io/blog/four-files-that-keep-a-long-agent-run-on-track)
- [Why did every AI tool start shipping “skills” at the same time?](https://loreto.io/blog/why-everyone-shipped-skills-at-once)
- [What breaks first when one person runs five agents instead of one?](https://loreto.io/blog/one-person-five-agents-what-breaks)
- [Can a compliance policy be a skill file?](https://loreto.io/blog/can-a-compliance-policy-be-a-skill-file)
- [What did the robot that beat Usain Bolt forget to learn?](https://loreto.io/blog/the-robot-that-beat-bolt-could-not-stop)
- [What is WebMCP, and how is it different from MCP?](https://loreto.io/blog/webmcp-vs-mcp-what-is-the-difference)
- [Why does my agent keep picking the wrong tool?](https://loreto.io/blog/why-your-agent-picks-the-wrong-tool)
- [What should an agent tool return when it fails?](https://loreto.io/blog/what-makes-an-agent-error-message-useful)
- [Should an agent be allowed to spend money on your site?](https://loreto.io/blog/should-an-agent-be-allowed-to-spend-money)