Skip to content

hktitan/kheskills

unversioned · 5168196f9e01MIT

Harshit Khemani's 19 agent skills: UI craft, design tokens, SVG, stickers, diagrams, sound, launch films, demos, packaging, the poteto pyramid, a content engine, and brains that answer from one source's public work with the source named. Start with khe-auto.

khe-auto

The mode that runs the rest. Reads a request, picks a playbook or the one or two skills that own it, says why in a line, then does the work in order. Covers every skill in this collection: the khe skills (UI craft, design tokens, SVG, diagrams, sound, launch films), the skills by Emil Kowalski and Benji Taylor linked beside them, Vercel's ui-sdk and agent skills, and the engineering skills of Lauren Tan (poteto) and Matt Pocock. Use when a request names no skill, spans several, or someone says "auto", "which skill", "where do I start", "what order", "list the skills", "make this better", "I need a launch video", "build this screen", "build it with ui-sdk", "we ship Thursday", "the agent keeps doing this". When the request already names a skill, use that skill.

khe-brain-benji

Answers design and product questions from Benji Taylor's published writing and public skills, with the piece named and linked. Gives the claim, its title and its URL; pressure-tests a design, a product decision or a piece of faux UI against his stated positions; says he has not written about it when the corpus is silent; flags where his pieces disagree or changed over time. Use when someone says "what would Benji say", "would Benji ship this", "ask Benji", "who said that", "where does that come from", "is that really his rule", "pressure-test this against Benji", or wants a claim about restraint, framing, motion, delight, agent feedback or rehearsal traced to its source. It reads and cites and never builds: to build faux product UI use stageshow, his own skill; to build or review an interface use khe-ui; to pick a skill use khe-auto. A reading by Harshit Khemani, not endorsed by Benji Taylor.

khe-brain-emil

Answers interface, motion and craft questions the way Emil Kowalski's published articles answer them, with the article named. Gives the claim, its title and its URL; pressure-tests a UI or a decision against his stated positions; says "he has not written about this" when the articles are silent; shows where two articles disagree. Harshit Khemani's reading of his public writing, not written or endorsed by him. Use when someone says "what would Emil say", "is this how Emil would do it", "ask Emil", "who said that", "where does that rule come from", "has Emil written about", "does Emil contradict himself", or wants a motion, easing, component or taste decision checked against a named source. To build or review a screen use khe-ui. To apply his rules to code use his own skills (emil-design-eng, animate, review-animations), which khe-auto routes to.

khe-brain-enrico

Answers how tech products are designed, engineered and built from what Enrico Tartarotti has said in the 88 long videos on his YouTube channel, with the video and the time named. Gives the claim, its video, its time and a link; pressure-tests a design, a product decision or a product plan against his stated positions; says the videos do not cover it when they are silent; shows both sides where two videos disagree or he changed his mind. Covers hidden details, psychology, waiting and motion, notifications and feeds, AI products, interfaces, devices, why products win, business models and building products. Use when someone says "what would Enrico say", "ask Enrico", "what does Enrico Tartarotti think about", "which video was that", "who said that", "pressure-test my design against Enrico". It reads and cites and never builds: to build or review a screen use khe-ui; for posts in your own voice khe-content-engine; to pick a skill khe-auto. A reading by Harshit Khemani, not endorsed by Enrico Tartarotti.

khe-brain-harshit

Answers and drafts the way Harshit Khemani thinks, from every post he made on X and LinkedIn up to 13 November 2026, with the post named and linked. Gives his position, the post it comes from and its date; drafts a post for khe.money in his current voice (lowercase, short lines, one idea a line, a plain hook, no hashtags) that rests only on positions he has posted, each cited under the draft; pressure-tests a plan against his stated views; says his posts do not cover it when they are silent; shows both sides where he changed his mind. Covers product and design, games and Smash&Clash, AI and agents, tools and workflow, building and startups, distribution, learning, work and age, and his own story. Use when someone says "what does Harshit Khemani think about", "ask Harshit's brain", "write a post in Harshit's voice", "draft a khe.money post", "would Harshit say this", "did Harshit post about", "which post was that". For a content engine across platforms use khe-content-engine; to pick a skill use khe-auto.

khe-brain-isb

Answers business questions from two sets of sources, with the piece named and linked: the isb-skills notes, a public set of 457 MBA concept files that their README says draw on the ISB PGP 2025-26 curriculum, and 153 session notes from the 2025 ivi programme at ISB, each naming its speaker. Gives the claim, its title and its URL; pressure-tests a plan or a model against the pieces; says so when they are silent; reports both sides where two disagree; marks figures as unsourced. Use when someone says "what do the ISB notes say", "ask ISB", "what did they teach at ivi", "pressure-test my plan", "which file says that", "where does that number come from", or wants a claim about valuation, strategy, pricing, startups, fundraising, negotiation, operations, leadership or technology traced to its source. It reads and cites and never builds or advises: to pick a skill use khe-auto. Harshit Khemani's reading of a public repository and of sessions he attended, not written or endorsed by ISB or any speaker.

khe-brain-varun

Answers startup, content, career and AI questions from what Varun Mayya has said in 18 public videos (13 Katas, 2 talks, 3 podcasts), with the video and the minute named. Gives the claim, its video, its time and a link that opens there; pressure-tests a plan, a pitch or a career decision against his stated positions; says the videos do not cover it when they are silent; shows both sides where he disagrees with himself or changed his mind. Covers distribution, bootstrapping, risk, hiring, leading teams, brand as world building, attribution, AI and jobs, belief, money and status, India, and the Aeos story. Use when someone says "what would Varun say", "ask Varun", "what does Varun Mayya think about", "which Kata was that", "who said that", "pressure-test my plan against Varun". It reads and cites and never drafts: for posts in your own voice use khe-content-engine; to pick a skill use khe-auto. A reading by Harshit Khemani, not endorsed by Varun Mayya.

khe-brain-yc

Answers startup questions from 88 public Y Combinator essays, posts and talks, with the person who said it named and the piece linked. Gives the claim, who said it, the title, the URL and the date; pressure-tests a plan, a deck or a decision against stated positions; says no piece read covers it when the corpus is silent; shows both sides where partners disagree or advice changed over the years. Use when someone says "what does YC say about", "what would Paul Graham say", "ask YC", "who said that", "is that really YC advice", "where does that come from", "pressure-test my seed plan", "do the partners agree on", or wants a claim about ideas, co-founders, MVPs, product-market fit, growth, pricing, sales, fundraising, SAFEs, runway or hiring traced to its source. It reads and cites and never advises or builds: to pick a skill use khe-auto; for another person's writing use khe-brain-emil or khe-brain-benji. A reading by Harshit Khemani, not endorsed by Y Combinator.

khe-content-engine

Builds and runs a content engine: a folder of small linked markdown nodes in your own project (a voice sampled from posts you wrote, audience segments, one node per platform you use, hooks, repurposing rules, content types, cadence), then turns one idea into a different draft for each platform and names the node behind every choice. Also reviews drafts against the engine and tunes the node that caused a wrong line. It drafts; a person posts. Use when someone says "build me a content engine", "turn this into posts for X and LinkedIn", "repurpose this", "write in my voice", "make a thread out of this", "why doesn't this sound like me", or wants one idea written for several platforms. For a launch film use khe-launch-film, for a product demo khe-demo, for stickers khe-sticker, for faux product UI stageshow (Benji Taylor). When unsure which skill owns a job, khe-auto.

khe-demo

Records a demo of the real running product: seeds its state, writes the flow as a script file, drives the actual web app or desktop app, and hands back a short clean take (mp4 or webm, plus a GIF for a README) in which one thing happens and every pointer move has a cause. Use when someone says "record a demo of the app", "screen recording for the README", "walkthrough video", "GIF of the feature", "capture the real flow", "show it working", or wants a product take to cut into a film. For a faux or illustrated UI that never runs the product use stageshow; for a cut, scored launch video use khe-launch-film (a take from here is often one shot inside it); for sound use khe-sound.

khe-design-tokens

Finds a project's design contract (a .design file or DESIGN.md), resolves which one wins, and threads its tokens through the UI as CSS variables, a Tailwind theme or component props. Also writes design-system docs an agent can follow and checks the contract for drift. Use when a repo has a .design or DESIGN.md, or someone says "use our tokens", "make it on-brand", "stop hardcoding colors", "the agent keeps inventing variants" or "document our design system". To design the scales themselves, review the result or finish a screen that still feels unfinished, use khe-ui.

khe-diagram

Decides whether a diagram earns its place over a sentence or a table, picks the surface by who touches the picture next, then draws a finished editorial figure or edits a live tldraw canvas, with connector, accent and budget rules that also serve as the review. Use when someone says "can you diagram this", "draw the architecture", "make a flowchart", "sketch this on a board", "my diagram looks like Mermaid", or names tldraw, a .tldraw file or diagram-design. For SVG markup and animation use khe-svg; for a live chart inside a product use khe-ui; for a leader line on a product shot use stageshow.

khe-launch-film

Plans, cuts and reviews a product launch video against 47 measured films: writes the brief and finds the one moment the film must show, names the register, then times the beats, on-screen type, UI demo, shot motion, seams and cuts, and judges a render at native fps against its own register's numbers. Use when someone wants a launch film, sizzle, teaser, README film, logo sting, landing-page loop or a 9:16 cut, says "review my launch video", "what should the video show", "it looks good but people don't get it", "it feels like a slideshow", "when should the name appear", "how fast should the prompt type", or hands over a reference film to measure. For the soundtrack once picture is locked use khe-sound; for motion inside the product use khe-ui; for a faux UI demo on a page use stageshow.

khe-package

Packages something you built for people who have never heard of it: picks the format (an explainer rendered from HTML, a click-through of one live component, a hover reel, a recording page), sends three ideas as stills before any render, renders one frame at a time on a paused clock at 60 fps with the short side at 2000 px or more, crops to the work with 40 to 60 px of page around it, moves your system's real cursor on a slight curve and holds it still while the motion plays, then writes the post's first line for a stranger, not for you. Use when someone says "package this", "make a video of this for X", "render it at 4k", "a clip for the launch post", "how should I show this off", "give me ideas for the announcement video", or "package my work". To record a whole product flow use khe-demo; for a cut, scored film khe-launch-film; for faux UI stageshow (Benji Taylor); for the posts on each platform khe-content-engine. When unsure, khe-auto.

khe-poteto-pyramid

Puts every correction you give a coding agent at the highest layer of Lauren Tan's poteto pyramid that can hold it (the codebase, then lints, types and CI, then verification, then a line of prose, with human review last), builds the fix there, proves the check fails on a real past mistake, and deletes the instruction it replaces. Also sweeps a repo's history for repeated mistakes, triages an AGENTS.md or CLAUDE.md into checks and judgment calls, and audits how far agents can be trusted to merge. Use when someone says "the agent keeps doing X", "where should this fix live", "turn this into a lint", "stop it writing this again", "our AGENTS.md is too long", "can we let agents merge", "all code is slop", or names the poteto pyramid. For UI review use khe-ui, for a recording of the app khe-demo, for design corrections turned into taste rules learn-taste (Emil Kowalski). When unsure, khe-auto.

khe-sound

Decides which moments in a product UI or a launch video earn a sound or a haptic, designs one material family, generates the files with ElevenLabs or an offline synth, and places each transient on the frame where the motion lands. Use when someone says "add a sound", "it needs a satisfying click", "should this make a sound?", "the click feels late", "the sounds feel stock", or wants UI sound effects, a mute toggle, vibration on the mobile web, or a soundtrack placed on a locked film. For the motion a sound lands on use khe-ui; for a film's register, beats and cuts use khe-launch-film.

khe-sticker

Designs, animates and packs stickers of every kind: flat die-cut, lettering, character, badge, isometric slab and reaction, drawn as one set with one outline and one palette, each with a single motion that ends on a still, then rendered to static and animated SVG, PNG and the WhatsApp and Telegram formats. Use when someone asks to "make a sticker", wants a "sticker pack", "WhatsApp stickers", "Telegram stickers", an "isometric sticker", a mascot or reaction sticker, or says "animate this sticker". For icons, logos and general SVG use khe-svg; for the palette contract use khe-design-tokens; when a sticker needs a sound use khe-sound.

khe-svg

Creates, cleans and animates SVG files: icons, illustrations, logo reveals, stroke draw-ons, path morphs, mascots, and flat video clips turned into one editable vector file, each on the token system and moved by the engine its home allows. Use when someone asks to make or fix an SVG, says "this icon won't recolor", "the export is 200 KB", "it rotates around the corner", "the morph jumps", or wants a logo, icon or mascot to move. For page transitions, gestures and timing beyond the file use khe-ui; for choosing an icon pack for a product use khe-ui; for stickers and sticker packs use khe-sticker; for diagrams use khe-diagram.

khe-ui

Interface craft with numbers: motion, typography, color and surface, layout, components, accessibility, mobile feel, AI tells, review, and the pass before shipping. Builds or judges a screen against exact values and ends a review with a verdict. Use when someone says "review this component", "is this well built", "it works but feels unfinished", "make it feel right", "this looks AI-generated", "which easing", "the page jumps", "is this accessible", "feels like a website on my phone", "are we ready to ship", or is building a screen, a form, a card or a page, or styling one built on Vercel's UI SDK. For an SVG use khe-svg; for sound use khe-sound; for a launch video use khe-launch-film; for a .design file or DESIGN.md use khe-design-tokens.