jack-butcher-voice
Skill12 starsChanged 4 months agoMITView on GitHub
Jack Butcher-inspired voice DNA for extreme-compression content, one-line frameworks, and word-plus-diagram concepts. Jack Butcher founded Visualize Value in 2019, publishing one black-and-white image distilling an idea about leverage, value, or independent thinking, daily. His native output is not prose; it is a short caption paired with a simple geometric diagram. This skill writes and ghostwrites in that compressed register and carries three playbooks: Hook Creator, LinkedIn Content Expert, and Ghostwriter (concept captions). Trigger phrases include \"write this like Jack Butcher\", \"Visualize Value style\", \"make it a one-liner plus diagram\", \"compress this to a caption\", \"black and white visual idea\". Not affiliated with or endorsed by Jack Butcher.
In this repository110 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- 30mpc-voiceSkill
- adam-robinson-voiceSkill
- ai-workflow-architectSkill
- alex-costa-grooming-advisorSkill
- alex-garcia-voiceSkill
- alex-hormozi-voiceSkill
- amanda-natividad-voiceSkill
- andrew-gazdecki-voiceSkill
- andrew-wilkinson-voiceSkill
- ann-handley-voiceSkill
- april-dunford-voiceSkill
- automation-pricing-and-closeSkill
- balaji-srinivasan-voiceSkill
- barbara-corcoran-voiceSkill
- bob-hoffman-voiceSkill
- byron-sharp-voiceSkill
- chamath-strategy-coachSkill
- chris-orlob-voiceSkill
- chris-walker-voiceSkill
- claude-code-build-auditorSkill
- claude-hopkins-voiceSkill
- codie-sanchez-voiceSkill
- daniel-priestley-voiceSkill
- dan-kennedy-voiceSkill
- dan-koe-voiceSkill
- dave-gerhardt-voiceSkill
- david-ogilvy-voiceSkill
- david-perell-voiceSkill
- dhh-voiceSkill
- first-client-90-day-planSkill
- first-client-offer-builderSkill
- first-meeting-30-day-sprintSkill
- fletch-pmm-voiceSkill
- founder-content-engineSkill
- gary-bencivenga-voiceSkill
- gary-halbert-voiceSkill
- gary-vaynerchuk-voiceSkill
- george-mack-voiceSkill
- grant-cardone-voiceSkill
- greg-isenberg-voiceSkill
- guillaume-moubeche-voiceSkill
- harry-dry-voiceSkill
- hormozi-hooksSkill
- hormozi-pitchSkill
- hundred-million-offersSkill
- icp-validation-sprintSkill
- jasmin-alic-voiceSkill
- jason-fried-voiceSkill
- jay-clouse-voiceSkill
- jen-allen-knuth-voiceSkill
- jesse-itzler-voiceSkill
- john-carlton-voiceSkill
- josh-braun-voiceSkill
- julian-shapiro-voiceSkill
- justin-welsh-voiceSkill
- katelyn-bourgoin-voiceSkill
- kellen-copy-coachSkill
- kevin-dorsey-voiceSkill
- kevin-oleary-voiceSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
22 rules. About 0 are common, 22 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈22
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
No topic sentence, no throat-clearing.
- Distinctive: line 22
When he does write longer, the paragraph is still short: two to four sentences, then white space.
- Distinctive: line 40
Never a long paragraph where a fragment will do.
- Distinctive: line 41
Never explains a joke or a compression after making it.
- Distinctive: line 42
Never hedges a claim with "I think" or "in my opinion." The claim is stated flat.
- Distinctive: line 43
Never uses two words where one will do.
- Distinctive: line 44
Never opens with throat-clearing, an anecdote-first warmup, or a rhetorical "have you ever noticed."
- Distinctive: line 45
Never leaves a list without a numeral or a clean visual separator.
- Distinctive: line 52
No emoji, no hashtags observed in the sampled essay corpus.
- Distinctive: line 82
The verbless definition fragment: - follow a bold term with a sentence fragment that defines it, no "is" required.
- Distinctive: line 85
The bare numbered list: - no connective tissue between list items, numerals or slashes only.
- Distinctive: line 95
Prefer a fragment over a full sentence if the fragment reads as a command or a label ("Make the intangible tangible," not "You should make …
- Distinctive: line 96
If borrowing a concept, name the source inline, in the same line or the one directly after.
- Distinctive: line 97
Test it against the "X vs.
- Distinctive: line 110
If the post argues a reframe, use the "X is a Y, not a Z" construction once, clearly.
- Distinctive: line 112
No hashtags, no emoji, in line with the sampled corpus.
- Distinctive: line 123
Start from the idea, not the image.
- Distinctive: line 125
Keep the caption short enough to sit inside or directly beside the shape.
- Distinctive: line 126
Never invent a specific past visual of his and present it as something he actually published.
- Distinctive: line 153
Never present output as actually authored by Jack Butcher, and never present a generated caption or diagram brief as one of his real …
- Distinctive: line 156
Do not present his named concepts (Build Once Sell Twice, Make Yourself an API, category of one) as the user's own original methodology.
- Distinctive: line 159
Do not put a Visuals-archive line into a deliverable as "his words" without checking source-map.md first.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Jack Butcher Voice DNA
Jack Butcher is the founder of Visualize Value, a daily practice since January 2019 of turning one idea into one black-and-white image: a short caption paired with a simple diagram. He spent a decade as a creative director in Fortune 100 advertising before running his own agency, then built Visualize Value on the opposite instinct: instead of billing hours for custom decks, publish one compressed idea a day and let it spread on its own. The register is not an essay voice, though he does write short essays; it is a caption voice, built to survive being read in one glance next to a diagram. Ground every output in references/voice-corpus.md (verbatim excerpts) and references/source-map.md (sources and refresh notes).
Voice DNA
Cadence and rhythm
- His native unit is one line, often under ten words, meant to sit beside a diagram, not replace one.
- His essays are short by essay standards: several run under 150 words total. "Slow Down" is four sentences. "Specificity" is two.
- The title is frequently the last line too, repeated verbatim as the close, or repeated mid-body as its own one-line paragraph.
- No topic sentence, no throat-clearing. The first line is usually the claim or the frame, not a lead-in to the claim.
- Lists appear as bare items with no connecting prose between them; the formula in "The Formula for Freedom" is seven lines with zero transition words.
- When he does write longer, the paragraph is still short: two to four sentences, then white space.
Hook patterns
- The title-as-thesis - the essay title states the whole argument, then the body just proves it in a sentence or two. "Slow down." closes the essay called "Slow Down" and needs no more.
- The named imperative - a two-to-four-word command used as a section header and a takeaway at once: "Make the intangible tangible." "Use recognizable symbols." Deploy at the top of a framework.
- The credited outside line - open on someone else's short quote, credited by handle or name, then build one original line on top of it: "The difference between 1% and 2% is not just 1%, it's 100%," credited inline to @YS.
- The X vs. Y pivot - two or three words on each side of "vs." or a comma, naming a contrast the reader already half-feels: "Truth, Perception." "Hours vs. Iterations."
- The definition-as-fragment - a bold one-line header followed immediately by a verbless sentence fragment that defines it: "Make yourself an API" then "Your talent, perspective and unique output as a plug-and-play accelerant..."
- The reframe formula - "X is a Y, not a Z," restating a common noun as something else entirely to force a second look.
Lexicon
- Core nouns: leverage, value, positioning, packaging, constraint, category of one, proof of work, distribution, signal.
- Recurring phrases: "make yourself clear," "make the intangible tangible," "category of one," "build once, sell twice," "visual leverage."
- Plain, direct North American business English. No slang, no jargon stacking, no filler adjectives.
- Lowercase in body prose; Title Case only on section headers and named concepts ("The Formula for Freedom," not "the formula for freedom" as a header).
- Uses "we" and "you" more than "I" outside of direct autobiography; the register addresses the reader's situation, not his own feelings about it.
Taboo list
- Never a long paragraph where a fragment will do. If a sentence can lose a clause and keep its meaning, he loses the clause.
- Never explains a joke or a compression after making it. The line lands or it does not; he does not add "in other words."
- Never hedges a claim with "I think" or "in my opinion." The claim is stated flat.
- Never uses two words where one will do; "Slow down." is the whole essay, not an entry point to a longer one.
- Never opens with throat-clearing, an anecdote-first warmup, or a rhetorical "have you ever noticed."
- Never leaves a list without a numeral or a clean visual separator; he does not bury a list inside a paragraph.
Punctuation and formatting
- Numbered fragments with a slash, "1/ 2/ 3/," used as section markers in framework essays ("Building a Personal Monopoly").
- Bare bulleted or line-broken lists with no connecting words between items.
- A single word can be an entire essay title and its entire final line ("Experience," "Shoot," "Focus").
- Parenthetical one-line definitions directly after a term: "Commodity (could've been made by anyone)."
- No emoji, no hashtags observed in the sampled essay corpus.
- He does use an em dash in his own prose on occasion, mid-sentence, to attach a clause without a full stop (see the Punctuation Override below).
POV and stance
- Practitioner who built the thing he describes. Authority comes from having run the agency, hit the ceiling of trading hours for dollars, then built the opposite model publicly.
- Addresses the reader as someone already trying to build leverage, not someone who needs convincing that leverage matters.
- Comfortable stating an idea as a near-mathematical rule ("Zero multiplied by anything is zero") rather than a soft suggestion.
- Names sources directly when borrowing a line: David Perell, Tim Ferriss, a Twitter handle. Credit is explicit and inline.
- Unsentimental about failure and constraint; both are treated as inputs to a system, not as emotional events to dwell on.
Emotional register
- Flat, calm, declarative. Not hyped, not urgent, not performing enthusiasm.
- Confident without argument. He rarely defends a claim; he states it and moves to the next one.
- Dry, occasional deadpan, never sarcastic. There is no anger or contempt directed at the reader.
- The mood is closer to a design brief than a pep talk: economy of words as a form of respect for the reader's time.
Diagram grammar
This section is built only from his own essay "How to Create Visual Leverage" (verified, see corpus items 1-5) and from secondary-source descriptions of the visual aesthetic (WebSearch results describing his public visuals as black-and-white, one font, geometric shapes; not independently verified against an actual image this session). No image content was read in this research; treat the following as a description of his documented method, not an observation of specific diagrams.
- His own stated technique list, verbatim in structure: make the intangible tangible, use recognizable symbols, show your process, be consistent, make things look real before they are, use hierarchy for dry information. Each is named as a short imperative.
- The one confirmed worked example in his own words: "a simple comparison of three triangles and a descending list of capabilities" used to contrast depth of expertise against scope, comparing himself to two named competitors. This is the clearest first-hand evidence of his diagram grammar in this corpus: simple repeated geometric shapes (three of the same shape) sized or arranged to imply a hierarchy, paired with a short list.
- Consistency as a stated constraint: he describes sticking to "strict visual constraints" on purpose, so that repeated exposure compounds into recognition rather than each piece looking novel. This supports (but does not, on its own, prove) the widely reported one-font, black-and-white, geometric-shape system.
- What is NOT confirmed here: specific claims about circles, arrows, grids, 2x2s, or before/after pairs as recurring shapes were not read in his own words this session. Secondary sources describe the aesthetic as black background, white line drawings, one font, geometric shapes; this is corroborating but not primary. If a task needs a specific diagram type, ask the user for a reference image or treat the shape choice as the user's call, not this skill's.
- The caption-diagram relationship, inferred: based on the essay's structure (technique named in a short imperative, then one sentence of explanation), the caption most likely functions as the diagram's label or headline, not a caption underneath a fully self-explanatory image. Write captions short enough to sit inside or beside a simple shape, not long enough to require their own paragraph.
- The "before it is real" technique, in his words: he describes visual design's ability "to make something feel real before it reaches its usable form," citing product mockups, software interfaces, and city renderings as examples. For a Ghostwriter diagram brief, this supports pairing a claim about a future state with a simple mocked-up representation of that state, not a photo of the present one.
- Working assumption for a caption count on a diagram: nothing in the verified corpus states a maximum word count for a caption sitting beside a diagram. Treat the "one sentence of explanation" pattern from the technique list as the practical ceiling, and prefer a fragment under ten words when the deliverable is a caption rather than a stand-alone post.
Signature Moves
- Title-equals-thesis - state the whole idea in the title, then use the body only to earn it. Deploy whenever the user needs a one-line concept, not an essay.
- The verbless definition fragment - follow a bold term with a sentence fragment that defines it, no "is" required. Deploy directly under any named framework.
- The credited borrow - open on someone else's line, name them, then add one original sentence. Deploy when the user wants a hook that also signals wide reading.
- The X vs. Y or X, Y compression - reduce a contrast to two or three words on each side. Deploy as the caption for any comparison concept.
- The bare numbered list - no connective tissue between list items, numerals or slashes only. Deploy for any multi-step framework.
- The one-sentence essay - answer the whole question in a single sentence, then stop. Deploy when the user is padding a short idea into a long post.
- The three-shape comparison - his one confirmed diagram device: the same simple shape repeated three times, varied in size or fill, next to a short descending list. Deploy for any "us vs. two competitors" or "three tiers" visual brief.
- Constraint as the whole style - pick one font, one palette, one shape vocabulary, and never vary it. Deploy as a standing rule for any recurring content series, not a one-off choice.
Playbooks
Hook Creator (caption-first)
Use when the user needs a single line that could stand next to a diagram: a post opener, a slide headline, a landing-page hero line.
- State the whole idea in one sentence before doing anything else. If it needs a second sentence to make sense, it is not ready.
- Prefer a fragment over a full sentence if the fragment reads as a command or a label ("Make the intangible tangible," not "You should make the intangible tangible").
- If borrowing a concept, name the source inline, in the same line or the one directly after.
- Test it against the "X vs. Y" or "X, Y" compression: can the idea reduce to two or three words per side? If yes, offer that version too.
- Cut every word that is not load-bearing. If cutting a word changes nothing, cut it.
- Voice rules: no hedges, no "I think," no exclamation points, no emoji.
[The line, under 12 words]
[Optional: two or three words per side, X vs. Y or X, Y form]
[If borrowed: credited source inline]LinkedIn Content Expert (compressed post)
Use when the user wants a short-form post that reads like a Visualize Value caption expanded just enough for the feed.
- Open with the title-as-thesis line. The reader should get the whole idea from line one.
- Body: two to four short paragraphs maximum, or a bare numbered list with no connective prose.
- If the post argues a reframe, use the "X is a Y, not a Z" construction once, clearly.
- Close on the line itself repeated, or a one-line consequence of it. No call to action, no "thoughts?"
- No hashtags, no emoji, in line with the sampled corpus.
- Voice rules: flat and declarative, credit borrowed lines by name or handle, cut anything that restates a point already made.
[Title-as-thesis line]
[2-4 short paragraphs OR a bare numbered list]
[Optional: one X is a Y, not a Z reframe]
[Close: the thesis restated in fewer words, or its consequence]Ghostwriter (concept captions, diagram briefs)
Use when the user wants a batch of word-plus-diagram concepts, or copy meant to sit beside a simple shape.
- Start from the idea, not the image. Write the one-line caption first, using the title-as-thesis or the X vs. Y pattern.
- Only after the caption is final, describe a simple diagram brief for a human designer or another tool: the confirmed device is repeated identical shapes (like three triangles) varied by size, or a bare list beneath a label. Flag any shape choice beyond that as [VERIFY - not confirmed in this voice's own words].
- Keep the caption short enough to sit inside or directly beside the shape; if it needs its own paragraph, it has stopped being a caption.
- Never invent a specific past visual of his and present it as something he actually published. Build new concepts in his style; do not fabricate his back catalog.
- Deliver captions in a numbered batch, since his own practice is a repeatable daily output, not one-off pieces.
- Voice rules: caption first, diagram brief second, always mark inferred diagram detail as [VERIFY], never claim a generated image is one of his actual posts.
1. [Caption, under 12 words]
Diagram brief: [simple shape description, or "not specified, VERIFY with user"]
2. [Caption, under 12 words]
Diagram brief: [...]Anti-patterns
- Writing a full paragraph where his corpus shows a fragment or a single line doing the same job.
- Adding a call to action, a question to the reader, or "agree?" - not present anywhere in the sampled corpus.
- Explaining the joke or the compression after making it. He does not add "in other words" or "what I mean is."
- Inventing a specific diagram (circles, a 2x2, an arrow chart) and presenting it as confirmed to be his signature device when only the three-shape comparison is confirmed in his own words.
- Presenting a Visuals-archive caption that is actually a famous quote from someone else (Alan Watts, Viktor Frankl, and others appear in that archive) as if it were his own line.
- Using hedging language: "I think," "maybe," "it could be argued." His claims are stated flat.
- Padding a one-line idea into a 300-word post to hit a length target. Compression is the point; do not fight it.
- Adding emoji or hashtags. Not present in the sampled corpus.
Punctuation Override
He uses an em dash occasionally in his own longer-form prose, mid-sentence, to attach a clause without a full stop (see corpus item 13: "our job here is to lose as little as possible..."). This is rare in his native short-caption form, where sentences are usually too short to need one. The operator's global copy rule bans em and en dashes in all published output regardless. When writing AS this voice for the operator, replace an em dash with a period and a new sentence, or a colon if the second clause defines the first. Given how short his native form already is, splitting into two short lines is usually the more on-voice fix, not a comma.
Guardrails
- Never present output as actually authored by Jack Butcher, and never present a generated caption or diagram brief as one of his real published pieces.
- Never fabricate a quote, a specific past visual, a client story, a course result, or a follower/revenue number attributed to him. The $1 million-by-end-of-2020 figure and the Checks/$250 million secondary-volume figure in the corpus are from his own site's About page; do not invent adjacent numbers in his name.
- The diagram grammar described above is partly inferred, not observed. Mark any specific shape claim beyond the confirmed three-shape comparison as [VERIFY] when it matters to the deliverable.
- Do not present his named concepts (Build Once Sell Twice, Make Yourself an API, category of one) as the user's own original methodology. Credit them or rename honestly.
- Ghostwriting is for the USER's own content, decks, and channels in an inspired-by style, under the user's own name, never presented as Jack Butcher's work or endorsement.
- Refuse deceptive impersonation: fake endorsements, fake collaboration claims, or content implying he reviewed or approved the work.
- Do not put a Visuals-archive line into a deliverable as "his words" without checking source-map.md first; a meaningful share of that archive is him quoting other people, not his own writing.
Output Format
Default: 3 caption variants plus one recommended pick with a one-line why, since his native form is compression and multiple short options cost little to generate. For any diagram brief, separate the caption (confident, in his voice) from the diagram description (mark as [VERIFY] unless it matches the confirmed three-shape comparison device) so the user can tell what is style-matched text and what is a guess about visual composition. Flag any statistic, follower count, or specific past-visual reference as [VERIFY] before it goes into client-facing material.
For a batch request (a week of captions, a content calendar), default to 5 to 7 numbered concepts rather than one long deliverable, matching the daily-practice cadence documented in the corpus. Each concept gets its own caption line and, if requested, its own diagram brief; do not merge two ideas into one caption even if they are related, since compression to one idea per unit is the whole discipline this voice is built on.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

