Get AI summaries of any video or article — Sign up free
Claude Skills - SOPs For Agents thumbnail

Claude Skills - SOPs For Agents

Sam Witteveen·
5 min read

Based on Sam Witteveen's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

SOP-style packaging is becoming a standard way to improve agent reliability by structuring inputs, steps, tool use, and outputs.

Briefing

Standard operating procedures are becoming a core building block for AI agents—not just a human workflow tool. The shift matters because LLM output quality depends heavily on the context and instructions provided, and SOP-style structures reduce variability by forcing inputs, step-by-step actions, tool use, and end-of-task verification into a repeatable pattern.

Anthropic’s newly announced “Claude skills” is the clearest example of this trend. Skills are packaged folders containing instructions, scripts, and resources that Claude can load only when relevant. Anthropic has also prepared pre-made skills for common office and document tasks such as working with Excel files, creating presentations, and formatting documents. The practical idea is that these packages act like reusable context bundles: instead of re-prompting from scratch, an agent can pull in the right instructions and assets at the moment they’re needed.

Claude skills aren’t limited to the Claude chat interface. They’re available through the API as well, where skills can be added to messages API requests via a dedicated endpoint, alongside Claude’s code execution tool for running tasks that depend on those skill assets. Anthropic describes skills as composable and portable—stackable across workflows and usable across Claude apps, Claude Code, and API-driven agent setups. That “load only when needed” behavior is central: it’s designed to keep the model’s context focused rather than bloated.

A key feature is a “skill creator” capability inside Claude settings. It provides interactive guidance to design a workflow, then generates the folder structure and a “skill MD” file that defines the skill’s name, description, steps, syntax, and tool usage. The transcript draws a parallel to how coding agents rely on custom instruction files (like agents.md-style patterns), but with the twist that the system can help author the SOP package itself.

The broader ecosystem is moving in the same direction. The transcript links Claude skills to Claude code plugins and Google’s Gemini CLI extensions, framing them as packaging mechanisms for agent context—whether through CLI commands, MCP servers, or other tool integrations. It also notes that foundation-model providers are diverging in their APIs to better support background tools and agent execution, including examples like Gemini API integrations (e.g., Google search) and agent configurations that connect skills with MCPs and backend code execution.

Finally, the transcript highlights practical considerations beyond functionality: token conservation best practices, the presence of a public repo of skills (including examples with code for packaging and validation), and licensing differences across skill packages (some Apache 2 licensed, others strictly copyrighted). It ends with a recommendation to try Claude skills directly—such as converting YouTube transcripts into LinkedIn posts—and to consider turning off training on skills if protecting proprietary SOPs is a priority. Overall, the direction is clear: chat platforms are evolving into app-like agent platforms, with “context window marketplaces” emerging around reusable SOP packages.

Cornell Notes

AI agents are increasingly built around SOP-style packages that standardize inputs, tool steps, and outputs to reduce variability. Anthropic’s “Claude skills” packages instructions, scripts, and resources into folders that Claude loads only when relevant, with pre-made skills for tasks like Excel work, presentations, and document formatting. Skills are composable and portable across Claude chat, Claude Code, and the API, where they can be attached to messages requests and used alongside code execution. A built-in “skill creator” helps generate the required folder structure and a “skill MD” file that defines the workflow. The same packaging approach is spreading across other ecosystems via plugins and CLI extensions, pointing toward marketplaces of reusable agent context.

Why do SOP-like structures matter more for agents than for humans?

LLM quality is tightly tied to the context provided. SOPs reduce variability by forcing consistent inputs and a repeatable sequence of steps. For agents, that means structuring the workflow into inputs → ordered steps → decision points → tool use → output → optional verification, so the model doesn’t improvise differently each run.

What exactly is a “Claude skill,” and when does Claude use it?

A Claude skill is a folder containing instructions, scripts, and resources. Claude loads the skill only when it’s relevant to the task at hand, which helps keep the context window focused instead of stuffing everything in every request.

How do Claude skills work across interfaces and the API?

Skills aren’t confined to the Claude chat UI. They can be used in Claude apps, Claude Code, and through the API by adding skills to messages API requests via a dedicated endpoint. The setup also pairs with Claude’s code execution tool to access skill folders and run tasks like document formatting or PDF handling.

What does the “skill creator” do?

In Claude settings, the “skill creator” provides interactive guidance to design a workflow and then generates the folder structure plus a “skill MD” file. That skill MD file directs the skill’s steps, syntax, and tool usage—similar in spirit to custom instruction files used by coding agents, but with automated scaffolding.

How does this relate to MCPs, plugins, and Gemini CLI extensions?

The transcript frames skills as part of a broader packaging pattern that also appears in MCP-related prompt/context layers. It also links Claude skills to Claude code plugins and Google’s Gemini CLI extensions, where the goal is the same: package reusable context and tool wiring so the model can act correctly when needed.

What practical issues show up when adopting skills—beyond functionality?

Token conservation is emphasized, including examples of extracting PDF content using relatively few tokens. There’s also attention to licensing (some skills Apache 2 licensed, others strictly copyrighted) and a suggestion to disable training on skills if proprietary SOPs must be protected.

Review Questions

  1. How does “load only when relevant” change the way skills should be designed compared with always-on prompt stuffing?
  2. What role does the “skill MD” file play in turning a workflow into an executable SOP package?
  3. Why might token conservation influence how a skill handles PDFs or other large inputs?

Key Points

  1. 1

    SOP-style packaging is becoming a standard way to improve agent reliability by structuring inputs, steps, tool use, and outputs.

  2. 2

    Claude skills are folders of instructions, scripts, and resources that Claude loads only when relevant to the task.

  3. 3

    Claude skills are usable across Claude chat, Claude Code, and the API, including via messages API requests and code execution.

  4. 4

    A built-in “skill creator” can generate the required folder structure and a “skill MD” file that defines workflow steps and tool syntax.

  5. 5

    Skills are designed to be composable and portable, enabling stacking of multiple skills in one workflow.

  6. 6

    The ecosystem is converging on similar packaging ideas across providers via plugins and CLI extensions, even as APIs diverge.

  7. 7

    Adoption requires attention to token efficiency, licensing terms, and whether skills should be excluded from model training.

Highlights

Claude skills package instructions and assets into folders that load only when relevant, aiming to keep context windows lean and task-focused.
The “skill creator” can turn an interactive workflow design into a structured skill folder plus a “skill MD” file that governs steps and tool usage.
Skills are positioned as reusable SOP units across Claude chat, Claude Code, and API-driven agents, with composability and portability as first-class goals.
Licensing varies across skill packages—some Apache 2 licensed, others strictly copyrighted—so reuse and redistribution may require careful review.

Topics

Mentioned

  • MCP