A Skill Is A Folder, Not A Prompt: What Anthropic Learned Running Hundreds Of Them

📊 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 a folder-based system called Skills, which encapsulate instructions, scripts, and knowledge. This approach improves consistency, onboarding, and organizational learning for AI agents.

Anthropic has revealed that its AI teams have transitioned from relying on prompts to using a new structure called Skills, which are folders containing instructions, scripts, and reference materials. This approach aims to make AI outputs more consistent, improve onboarding, and create an evolving library of organizational knowledge. The shift represents a fundamental change in how AI agents are built and maintained, emphasizing durable, reusable assets over ad-hoc prompting.

In a detailed write-up from a Claude Code engineer, Anthropic explained that a Skill is not merely a saved prompt but a folder that can include instructions, reference documents, scripts, templates, and configuration data. The agent can discover and execute the contents of this folder, enabling more complex behaviors and institutional knowledge retention.

This method transforms the traditional prompt-based approach into a structured, versioned asset that organizations can develop, share, and improve over time. Anthropic’s internal experience shows that Skills help standardize output, streamline onboarding, and compound organizational knowledge, turning repetitive tasks into reliable, reusable modules. The company estimates that investing a week of engineering effort into perfecting a Skill can yield significant long-term benefits.

Anthropic categorizes its Skills into nine types, including library references, product verification, data analysis, automation, code scaffolding, review processes, deployment, runbooks, and infrastructure operations. Among these, verification Skills—used to check work quality—are considered most valuable, as they directly improve output reliability. The emphasis is on building Skills that catch mistakes and enforce standards, rather than just generating code or content.

At a glance
reportWhen: announced March 2024
The developmentAnthropic shared insights from its internal experience running hundreds of Skills, emphasizing a folder-based structure over prompts to enhance AI operational capabilities.
A Skill Is a Folder, Not a Prompt — Insights
AI Dispatch · Insights · 1 July 2026

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.

✕ The misconception

“A Skill is just a clever markdown prompt you save in a file.”

✓ What it actually is

A folder the agent can discover, read & run — instructions, scripts, references, templates, config & on-demand hooks.

Anatomy of a Skill — the file system is context engineering
my-skill/the unit you share & version
├─ SKILL.mdroot instructions + a description written for the model (its trigger)
├─ references/deep detail pulled in only when needed — progressive disclosure
├─ scripts/real code, so the agent composes instead of rebuilding boilerplate
├─ assets/templates & files to copy into the output
├─ config.jsonsetup the agent asks for if it’s missing (e.g. which Slack channel)
└─ hooks + memoryon-demand guardrails + an append-only log so it remembers
Why it matters: the folder itself is the knowledge base. The agent reads the root, then reaches deeper only when the task demands it — the same way you’d hand a new hire a one-pager that points to the detailed docs.
The nine types — a gap-analysis map for your own library
1Library / API reference
2Product verification ★ top impact
3Data fetching & analysis
4Business-process automation
5Code scaffolding & templates
6Code quality & review
7CI/CD & deployment
8Runbooks
9Infrastructure operations
By Anthropic’s own measurement, verification Skills — the ones that check the work — moved output quality the most. If you build one category well, build that one.
The craft — what separates a good Skill from a useless one
Gotchas = highest-signal section Describe for the model, not humans (it’s the trigger) Don’t state the obvious Ship scripts, not just prose On-demand guardrail hooks (/careful, /freeze) Let it remember (log / SQLite) Don’t railroad — leave room to adapt
The take

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.

Source: “Lessons from building Claude Code: How we use skills,” Thariq Shihipar (Anthropic), Claude blog, 3 June 2026. Categories, examples & measured claims are Anthropic’s; framing is the author’s. Docs: code.claude.com/docs/en/skills.
thorstenmeyerai.com

Implications for AI Development and Organizational Knowledge

This shift from prompts to folder-based Skills signifies a move toward more robust, maintainable AI systems. By encapsulating tribal knowledge, guardrails, and tools within Skills, organizations can achieve more consistent outputs and reduce onboarding time for new team members. The approach also allows companies to build a library of evolving best practices, turning routine procedures into reliable, sharable assets. This method addresses long-standing challenges in AI deployment—such as variability, knowledge silos, and fragility—by creating a durable, version-controlled repository of operational procedures.

For businesses, adopting Skills could mean more predictable AI behavior, easier scaling, and a clearer way to capture and improve internal expertise. The focus on verification Skills highlights the importance of quality control, which can significantly impact the reliability of AI-driven processes and outputs.

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

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 Building: Evolution in AI Practice

Traditional AI development often relies on prompt engineering—crafting specific instructions for the model to generate desired outputs. This approach, while flexible, suffers from inconsistency and difficulty in scaling. Anthropic’s internal experiments with hundreds of Skills demonstrate a different paradigm: structuring knowledge and procedures into organized folders that serve as reusable assets.

This development aligns with broader trends in AI toward modularity and robustness. Previously, organizations have struggled with prompt fragility, where small changes or misunderstandings lead to failures. The Skills approach seeks to address this by embedding instructions, scripts, and guardrails directly into organized containers that can be versioned, shared, and refined over time.

Anthropic’s experience suggests that Skills can be developed incrementally, starting with simple templates and evolving into comprehensive operational modules. This mirrors software engineering practices and emphasizes the importance of institutional memory in AI deployment.

“Shifting from prompts to folder-based Skills fundamentally changes how organizations build and maintain AI agents, making them more reliable and easier to scale.”

— Thorsten Meyer, AI researcher

Unclear Aspects of Skill Implementation and Adoption

It remains unclear how widely and quickly organizations will adopt the Skills framework outside of Anthropic. The practical challenges of integrating such structures into existing workflows, tools, and AI systems are still being evaluated. Additionally, the long-term impact on AI performance, especially in dynamic or complex environments, has not been fully demonstrated yet. The scalability and maintenance of Skills libraries as they grow in size and complexity also require further observation.

Next Steps for Broader Adoption and Validation

Organizations interested in this approach will likely begin experimenting with Skills in controlled settings, measuring improvements in consistency and onboarding. Anthropic plans to share more detailed case studies and best practices, encouraging industry-wide adoption. Future developments may include tooling improvements for managing Skills, automated validation, and integration with existing AI platforms. Monitoring how Skills perform in real-world deployments over the coming months will be key to understanding their full potential.

Key Questions

How is a Skill different from a prompt?

A Skill is a structured folder containing instructions, scripts, and reference materials, serving as a reusable organizational asset. A prompt is a single instruction or question sent to the AI. Skills provide durability and consistency beyond ad-hoc prompts.

What benefits do Skills offer over traditional prompt engineering?

Skills improve output consistency, onboarding efficiency, and organizational knowledge retention. They enable versioning, sharing, and systematic refinement, reducing fragility and scaling challenges.

Are Skills applicable to all types of AI tasks?

While initially focused on coding and operational procedures, the Skills framework can be adapted to various domains requiring reliable, repeatable AI outputs, including customer support, data analysis, and automation.

What are the main challenges in implementing Skills?

Key challenges include integrating Skills into existing workflows, managing growing libraries, and ensuring that Skills remain up-to-date and relevant as organizational needs evolve.

Will this approach replace prompt engineering entirely?

Not immediately. Skills complement prompt engineering by providing a more durable, structured foundation for complex and repetitive tasks. Over time, they may reduce reliance on ad-hoc prompting for routine operations.

Source: ThorstenMeyerAI.com

You May Also Like

Lifehacker Deals Live Blog: The Best Tech Sales, All in One Place

Stay updated with Lifehacker’s live blog showcasing the best current tech deals across gadgets, accessories, and more, curated daily for smart shoppers.

AI output review queue for customer support macros

Support teams are testing a new AI macro review queue to ensure policy compliance and appropriate tone before publication, aiming to improve support quality.

7 Best Wireless Smartwatches for Prime Day Deals in 2026

Discover the best wireless smartwatches to buy during Prime Day 2026, including Apple, Garmin, and budget options, with detailed insights on features and deals.

Operational SOP drift detector for franchise operators

A new SOP drift detection tool for multi-location franchise operators is being tested to identify local procedure changes and maintain consistency.