“Unusual was obliged to invent a lot of the magic he did, working from basic ideas and half-remembered tales from previous books.”
— Susanna Clarke, Jonathan Unusual & Mr Norrell
Fairy tales, myths, and fantasy fiction are stuffed with magic spells. You say “abracadabra” and one thing profound occurs.1 Say “open sesame” and the door swings open.
It seems that that is additionally a helpful metaphor for what occurs with giant language fashions.
I first bought this concept from David Griffiths’s O’Reilly course on utilizing AI to spice up your productiveness. He gave a easy instance. You’ll be able to inform ChatGPT “Set up my activity checklist utilizing the Eisenhower four-sided field.” And it simply is aware of what to do, even if you happen to your self know nothing about Basic Dwight D. Eisenhower’s method to resolution making. David then suggests his college students as a substitute attempt “Set up my activity checklist utilizing Getting Issues Performed,” or simply “Use GTD.” Every of these phrases is shorthand for techniques of thought, practices, and conventions that the mannequin has realized from human tradition.
These are magic phrases. They’re magic not as a result of they do one thing unworldly and sudden however as a result of they’ve the ability to summon patterns which were encoded within the mannequin. The phrases act as keys, unlocking context and even complete workflows.
All of us use magic phrases in our prompts. We are saying one thing like “Replace my resume” or “Draft a Substack put up” with out considering how a lot detailed prompting we’d must do to create that output if the LLM didn’t already know the magic phrase.
Each discipline has a specialised language whose phrases are identified solely to its initiates. We could be fanciful and faux they’re magic spells, however the actuality is that every of them is known as a sort of fuzzy operate name to an LLM, bringing in a physique of context and unlocking a set of behaviors and capabilities. Once we ask an LLM to jot down a program in Javascript reasonably than Python, we’re utilizing certainly one of these fuzzy operate calls. Once we ask for output as an .md file, we’re doing the identical. In contrast to a operate name in a conventional programming language, it doesn’t all the time return the identical consequence, which is why builders have a possibility to reinforce the magic.
From Prompts to Purposes
The following gentle bulb went off for me in a dialog with Claire Vo, the creator of an AI utility referred to as ChatPRD. Claire spent years as a product supervisor, and as quickly as ChatGPT turned out there, started utilizing it to assist her write product requirement paperwork or PRDs. Each product supervisor is aware of what a PRD is. When Claire prompted ChatGPT to “write a PRD,” it didn’t want an extended preamble. That one acronym carried a long time {of professional} apply. However Claire went additional. She refined her prompts, improved them, and taught ChatGPT the way to assume like her. Over time, she had educated a system, not on the mannequin degree, however on the degree of context and workflow.
Subsequent, Claire turned her workflow right into a product. That product is a software program interface that wraps up various associated magic phrases right into a helpful package deal. It controls entry to her personalized magic spell, so to talk. Claire added detailed prompts, integrations with different instruments, entry management, and a complete lot of conventional programming in a next-generation utility that makes use of a mixture of conventional software program code and “magical” fuzzy operate calls to an LLM. ChatPRD even interviews customers to be taught extra about their targets, customizing the applying for every group and use case.
Claire’s quickstart information to ChatPRD is a good instance of what a magic-word (fuzzy operate name) utility seems like.
You can too see how magic phrases are crafted into magic spells and the way these spells are even a part of the structure of purposes like Claude Code via the explorations of builders like Jesse Vincent and Simon Willison.
In “How I’m Utilizing Coding Brokers in September, 2025,” Jesse first describes how his claude.md file supplies a base immediate that “encodes a bunch of course of documentation and guidelines that do a reasonably good job conserving Claude on monitor.” After which his workflow calls on a bunch of specialised prompts he has created (i.e., “spells” that give clearer and extra personalised that means to particular magic phrases) like “brainstorm,” “plan,” “architect,” “implement,” “debug,” and so forth. Notice how inside these prompts, he could use further magic phrases like DRY, YAGNI, and TDD, which check with particular programming methodologies. For instance, right here’s his planning immediate (boldface mine):
Nice. I want your assist to jot down out a complete implementation plan.
Assume that the engineer has zero context for our codebase and questionablestyle. doc all the things they should know. which recordsdata to the touch for everyactivity, code, testing, docs they may have to verify. the way to take a look at it.givethem the entire plan as bite-sized duties. DRY. YAGNI. TDD. frequent commits.
Assume they're a talented developer, however know virtually nothing about ourtoolset or downside area. assume they do not know good take a look at designveryeffectively.
please write out this plan, in full element, into docs/plans/
However Jesse didn’t cease there. He constructed a venture referred to as Superpowers, which makes use of Claude’s lately introduced plug-in structure to “give Claude Code superpowers with a complete abilities library of confirmed methods, patterns, and instruments.” Asserting the venture, he wrote:
Expertise are what give your brokers Superpowers. The primary time they actually popped up on my radar was just a few weeks in the past when Anthropic rolled out improved Workplace doc creation. When the characteristic rolled out, I went poking round a bit – I requested Claude to inform me all about its new abilities. And it was solely too joyful to dish…. [Be sure to follow this link! – TOR]
One of many first abilities I taught Superpowers was The best way to create abilities. That has meant that after I wished to do one thing like add git worktree workflows to Superpowers, it was a matter of describing how I wished the workflows to go…after which Claude put the items collectively and added a pair notes to the present abilities that wanted to clue future-Claude into utilizing worktrees.
After studying Jesse’s put up, Simon Willison did a bit extra digging into the unique doc dealing with abilities that Claude had introduced and that had sparked Jesse’s brainstorm. He famous:
Expertise are extra than simply prompts although: the repository additionally contains dozens of pre-written Python scripts for performing frequent operations.
pdf/scripts/fill_fillable_fields.py for instance is a customized CLI device that makes use of pypdf to seek out after which fill in a bunch of PDF type fields, specified as JSON, then render out the ensuing mixed PDF.
It is a actually refined set of instruments for doc manipulation, and I like that Anthropic have made these seen—presumably intentionally—to customers of Claude who know the way to ask for them.
You’ll be able to see what’s taking place right here. Magic phrases are being enhanced and given a extra rigorous definition, and new ones are being added to what, in fantasy tales, they name a “grimoire,” or e book of spells. Microsoft calls such spells “metacognitive recipes,” an exquisite time period that ought to get extensively adopted, although on this article I’m going to stay with my fanciful analogy to magic.
At O’Reilly, we’re working with a really totally different set of magic phrases. For instance, we’re constructing a system for exactly focused competency-based studying, via which our clients can skip what they already know, grasp what they want, and show what they’ve realized. It additionally provides company studying system managers the power to assign studying targets and to measure the ROI on their funding.
It seems that there are dozens of studying frameworks (and that’s itself a magic phrase). Within the design of our personal specialised studying framework, we’re invoking Bloom’s taxonomy, SFIA, and the Dreyfus Mannequin of Ability Acquisition. However when a buyer says, “We love your method, however we use LTEM,” we are able to invoke that framework as a substitute. Each company buyer additionally has its personal specialised tech stack. So we’re exploring the way to use magic phrases to let no matter we construct adapt dynamically not solely to our finish customers’ studying wants however to the tech stack and to the educational framework that already exists at every firm.
That might be a nightmare if we needed to help dozens of various studying frameworks utilizing conventional processes. However the issue appears way more tractable if we’re capable of invoke the correct magic phrases. That’s what I imply after I say that magic phrases are an important constructing block within the subsequent technology of utility programming.
The Structure of Magic
Right here’s the essential factor: Magic isn’t arbitrary. In each mythic custom, it has construction, self-discipline, and price. The magician’s energy depends upon understanding the correct phrases, pronounced in the correct approach, with the correct intent.
The identical is true for AI techniques. The effectiveness of our magic phrases depends upon context, grounding, and suggestions loops that give the mannequin dependable details about the world.
That’s why I discover the rising ecosystem of AI purposes so fascinating. It’s about offering the correct context to the mannequin. It’s about defining vocabularies, workflows, and roles that expose and make sense of the mannequin’s skills. It’s about turning implicit cultural information into express techniques of interplay.
We’re solely firstly. However simply as early programmers realized to construct structured software program with out spelling out actual machine directions, right now’s AI practitioners are studying to construct structured reasoning techniques out of fuzzy language patterns.
Magic phrases aren’t only a poetic picture. They’re the syntax of a brand new sort of computing. As individuals turn into extra snug with LLMs, they are going to go across the magic phrases they’ve realized as energy consumer tips. In the meantime, builders will wrap extra superior capabilities round present magic phrases and maybe even train the fashions new ones that haven’t but had the time to accrete enough that means via large utilization within the coaching set. Every utility will likely be constructed round a shared vocabulary that encodes its area information. Again in 2022, Mike Loukides referred to as these techniques “formal casual languages.” That’s, they’re spoken in human language, however do higher whenever you apply a little bit of rigor.
And at the very least for the foreseeable future, builders will write “shims” between the magic phrases that management the LLMs and the extra conventional programming instruments and methods that interface with present techniques, a lot as Claire did with ChatPRD. However finally we’ll see true AI to AI communication.
Magic phrases and the spells constructed round them are solely the start. As soon as individuals begin utilizing them in frequent, they turn into protocols. They outline how people and AI techniques cooperate, and the way AI techniques cooperate with one another.
We are able to already see this taking place. Frameworks like LangChain or the Mannequin Context Protocol (MCP) formalize how context and instruments are shared. Groups construct agentic workflows that rely upon a standard vocabulary of intent. What’s an MCP server, in any case, however a mapping of a fuzzy operate name right into a set of predictable instruments and providers out there at a given endpoint?
In different phrases, what was as soon as a set of magic spells is changing into infrastructure. When sufficient individuals use the identical magic phrases, they cease being magic and begin being requirements—the constructing blocks for the following technology of software program.
We are able to already see this development with MCP. There are three distinct sorts of MCP servers. Some, like Playwright MCP, are designed to make it simpler for AIs to interface with purposes initially designed for interactive human use. Others, just like the GitHub MCP Server, are designed to make it simpler for AIs to interface with present APIs, that’s, with interfaces initially designed to be referred to as by conventional packages. However some are designed as a frontend for a real AI-to-AI dialog. Different protocols, like A2A, are already optimized for this third use case.
However in every case, an MCP server is known as a dictionary (or in magic phrases, a spellbook) that explains the magic phrases that it understands and the way to invoke them. As Jesse Vincent put it to me after studying a draft of this piece:
The half that feels essentially the most like magic spells is the half that almost all MCP authors do extremely poorly. Every device has a “description” discipline that tells the LLM how you employ the device. That description discipline is learn and internalized by the LLM and modifications the way it behaves. Anthropic are significantly good at device descriptions and most everyone else, in my expertise, is…much less good.
In some ways, publishing the prompts, device descriptions, context, and abilities that add performance to LLMs could also be a extra essential frontier of open supply AI than open weights. It’s essential that we deal with our enhancements to magic phrases not as proprietary secrets and techniques however as shared cultural artifacts. The extra open and participatory our vocabularies are, the extra inclusive and artistic the ensuing ecosystem will likely be.
Footnotes
- Whereas typically related right now with stage magic and cartoons, this magic phrase was apparently used from Roman occasions as a therapeutic spell. One proposed etymology means that it comes from the Aramaic for “I create as I converse.”
