📊 Full opportunity report: A Skill Is A Folder, Not A Prompt: What Anthropic Learned Running Hundreds Of Them on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
Anthropic has shifted from viewing AI skills as prompts to treating them as folders containing instructions, scripts, and knowledge. This approach enhances consistency, onboarding, and organizational learning, marking a significant change in AI operational strategies.
Anthropic has announced a new approach to building AI capabilities, defining Skills as folders, not prompts. This shift allows organizations to create reusable, structured containers of instructions, code, and knowledge that improve consistency, onboarding, and institutional memory. The development was shared through a detailed write-up by a Claude Code engineer, highlighting practical lessons learned from deploying hundreds of Skills internally.
According to Anthropic, a Skill is not merely a saved prompt but a folder containing instructions, reference documents, scripts, templates, data, and configuration. The agent can discover, read, and execute the contents within this folder, enabling more durable and reliable AI behavior.
This conceptual reframe is significant for both technical and business users. It means that output consistency improves, onboarding becomes more efficient, and Skills can evolve and compound over time. Anthropic emphasizes that investing engineer time into refining Skills can yield long-term organizational assets rather than one-off prompts.
Anthropic identified nine core categories of Skills, including library references, product verification, data analysis, automation, code scaffolding, review, deployment, runbooks, and infrastructure operations. The most impactful, according to the company, is verification — the Skills that check and validate output quality, which directly enhances reliability.
A Skill is a folder, not a prompt
Anthropic published what it learned running hundreds of Skills across its own engineering org. Read as a business memo, the point is bigger than a coding trick: this is how ad-hoc prompting becomes durable institutional capability — the SOPs your agents actually follow, versioned and shared.
“A Skill is just a clever markdown prompt you save in a file.”
A folder the agent can discover, read & run — instructions, scripts, references, templates, config & on-demand hooks.
The knowledge of how your organization actually operates can be captured, versioned, shared & executed — and the thing capturing it is a humble folder with a script and a gotchas list inside. For the builder, that’s context engineering with real tools attached. For whoever owns the budget, it’s the difference between AI that starts from zero every morning and an asset that compounds. Caveats: best practices are still evolving, checked-in Skills cost context, and curation beats accumulation. Start with one Skill, one gotcha, and the category that catches your mistakes.
Transforming AI Capabilities into Organizational Assets
This approach shifts how companies develop and maintain AI systems, emphasizing structured, reusable units over ephemeral prompts. It enables organizations to standardize processes, reduce onboarding time, and build a growing library of institutional knowledge. The focus on verification Skills highlights the importance of quality control in AI deployment, which is critical for enterprise adoption and safety.
Google Apps Script for Workflow Automation: Build Real-World Business Automation Systems Using AI-Assisted Learning, Triggers, APIs, and Enterprise Workflow Design
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Prompt Engineering to Modular Skill Development
Until now, most teams relied on prompt engineering—crafting specific instructions for each task. Anthropic’s internal experiments with hundreds of Skills demonstrated that packaging knowledge into folders yields more durable and scalable AI behaviors. The concept aligns with broader trends toward modular AI components, but Anthropic’s emphasis on structured folders as containers is a notable evolution.
Anthropic’s methodology is rooted in practical experience, with the company investing significant engineering effort into refining Skills, sometimes dedicating a full engineer-week per category. This contrasts with traditional prompt-based approaches, which are often ad-hoc and less maintainable.
“Viewing Skills as folders containing instructions and scripts fundamentally changes how organizations can build reliable AI systems.”
— Thorsten Meyer, AI researcher
Knowledge Engineering Tools and Techniques for AI Planning
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Unclear How Widespread Adoption Will Be
It is not yet clear how quickly other organizations will adopt this folder-based approach or how it will scale outside Anthropic’s internal environment. The effectiveness of Skills in diverse enterprise contexts remains to be proven, and there is ongoing debate about the best practices for structuring these folders and scripts.
AI instruction and script folders
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Next Steps for Industry Adoption and Standardization
Organizations interested in this approach should evaluate their current AI workflows and consider developing Skills as structured folders. Anthropic plans to share more detailed best practices and tools to facilitate broader adoption. Industry-wide, we may see a shift toward standardized Skill frameworks, especially for critical tasks like verification and automation, to improve reliability and maintainability.
Verification of Autonomous Systems
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
How does treating Skills as folders improve AI reliability?
By bundling instructions, scripts, and knowledge in a structured container, Skills enable consistent execution and easier updates, reducing errors caused by prompt variability.
Can this approach be applied outside of Anthropic?
Yes, the concept of modular, containerized knowledge units can be adopted by other organizations, though implementation details may vary based on infrastructure and use cases.
What are the main categories of Skills identified?
Anthropic identified nine categories, including library references, product verification, data analysis, automation, code scaffolding, review, deployment, runbooks, and infrastructure operations.
What remains unclear about this approach?
It is uncertain how scalable and effective this method will be across different industries and AI applications, and how organizations will structure and maintain large Skills libraries over time.
Will this change how AI models are trained or just how they are used?
This approach primarily affects operational procedures and how AI behaviors are managed and maintained, rather than the core training of models.
Source: ThorstenMeyerAI.com
