📊 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 using prompts to defining ‘Skills’ as folders containing instructions, scripts, and assets. This approach enhances AI consistency, onboarding, and organizational knowledge retention. The company ran hundreds of these Skills internally, demonstrating a new standard for AI operational procedures.
Anthropic has revealed a significant shift in AI development methodology: it now treats Skills not as prompts but as folders containing instructions, scripts, and reference assets. This approach, tested across hundreds of internal applications, aims to create more durable, consistent, and scalable AI operational procedures, marking a departure from traditional prompt engineering.
In a detailed write-up from a Claude Code engineer, Anthropic explained that a Skill is fundamentally a folder — a container for instructions, reference documents, scripts, templates, and configuration data. Unlike prompts, which are simple text instructions, Skills encapsulate complex, reusable organizational knowledge that AI agents can discover, read, and execute. This reframing allows for more durable and shareable operational routines, akin to institutional assets rather than ephemeral notes.
Anthropic’s internal experiments involved running hundreds of Skills across various engineering tasks, forming a library that covers nine categories: library and API references, product verification, data fetching, business-process automation, code scaffolding, code review, CI/CD deployment, runbooks, and infrastructure operations. The company found that the most valuable Skills focus on verification — checking outputs to prevent mistakes — which significantly improved output quality and reliability.
Technical lessons emphasize that effective Skills avoid restating obvious information and instead focus on non-obvious, specific knowledge. The description of each Skill acts as a trigger for the agent, matching user requests with the appropriate Skill based on natural language triggers, including internal slang. Bundling real code and helper functions within Skills further enhances their utility and 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.
Implications for AI Development and Organizational Knowledge
This new approach to Skills represents a shift toward more durable, scalable, and organizationally integrated AI routines. By encapsulating institutional knowledge into folders, companies can improve output consistency, accelerate onboarding, and build a growing library of operational assets. For organizations deploying AI at scale, this method offers a way to embed best practices directly into the AI’s routines, reducing reliance on ad-hoc prompting and manual instruction updates.
Anthropic’s internal findings suggest that investing engineering effort into refining Skills, especially verification Skills, yields tangible improvements in output quality. This approach also fosters a culture of continuous improvement, as Skills evolve through iterative refinement and documentation of edge cases. Ultimately, this shift could influence how enterprises standardize AI workflows and maintain institutional memory within their AI systems.
Create a Podcast with AI (No Experience Needed) : A Step-by-Step Guide to Planning, Scripting, Recording, Editing, and Launching a Podcast Using ChatGPT, AI Tools, and Automation
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Prompt Engineering to Asset-Based AI Routines
Traditionally, AI prompt engineering has involved crafting specific text instructions to guide model behavior. This method, while flexible, often results in fragile routines that require constant re-tuning. Anthropic’s move to treat Skills as folders containing instructions and code is a response to these limitations, aiming for more durable and reusable AI routines.
Prior to this development, many organizations relied on prompt tuning or manual scripting, which posed challenges in scaling and maintaining consistency. Anthropic’s internal experimentation with hundreds of Skills demonstrates a shift toward embedding organizational knowledge directly into the AI’s operational framework, akin to creating a library of best practices and guardrails.
This approach aligns with broader industry trends toward modular, asset-based AI systems that can be versioned, shared, and refined over time, reducing the need for repeated prompt re-engineering and manual intervention.
“Treating Skills as folders containing instructions, scripts, and assets fundamentally changes how we build and maintain AI routines. It turns ad-hoc prompts into durable organizational assets.”
— Thorsten Meyer, AI engineer at Anthropic
AI-READY SPECIFICATIONS: Write Clear Requirements That Help AI Agents Build Better Software, Products, and Business Documents with Less Rework
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Unclear Aspects of Skills Implementation and Adoption
While Anthropic’s internal results are promising, it is not yet clear how easily this approach can be adopted by other organizations or scaled across different AI use cases. Details about the tooling, standardization, and integration processes remain under development, and the long-term maintenance of Skills libraries is still being evaluated.
It is also uncertain whether this method will outperform prompt-based techniques in all scenarios, especially in highly dynamic or creative tasks. The impact on AI transparency and interpretability is also still being studied.
AI Excel & Spreadsheet Automation for Accountants: Copy-Paste Prompts, Templates, and AI Workflows to Save Time on Formulas, Reconciliation, Data Cleanup, … Tasks (AI Systems for Accountants Book 6)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Next Steps for Broader Adoption and Refinement
Anthropic plans to continue refining its Skills framework, exploring automation tools for cataloging, versioning, and updating Skills at scale. The company is also considering sharing its methodology with the broader AI community through open-source initiatives or industry collaborations.
Other organizations are likely to experiment with similar asset-based routines, testing whether Skills can replace prompt engineering in their workflows. Future developments may include standardized formats for Skills, better tooling, and integration with enterprise AI platforms.
Neiko 03029A 1/2" Scaffold Ratchet Wrench | 9.5” Length | 36 Tooth Hammer Tip Head | Includes 7/8” 6 Point Deep Socket | Cr-V Steel …
THE RIGHT TOOL: Extremely rugged specialty wrench for easier assembly and disassembly of scaffolding and other industrial-construction equipment
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is a ‘Skill’ in Anthropic’s new framework?
A Skill is a folder containing instructions, reference documents, scripts, templates, and configuration data that an AI agent can discover, read, and execute. It is a reusable organizational asset rather than a simple prompt.
How does treating Skills as folders improve AI performance?
It enhances consistency, scalability, and reliability by encapsulating complex, non-obvious organizational knowledge that can be versioned and shared across teams, reducing ad-hoc prompt tuning.
What categories of Skills did Anthropic identify?
Anthropic classified Skills into nine categories, including verification, data analysis, automation, code scaffolding, and infrastructure operations, focusing on operational and quality assurance tasks.
Can this approach be adopted by other organizations?
While promising, adoption depends on tooling, integration, and organizational readiness. Anthropic is exploring ways to share its methodology, but broader industry uptake remains to be seen.
Source: ThorstenMeyerAI.com
