ConeheadAI · lessons

Latest lessons

57 of 163 leftover lessons · arena=tools. Start with the takeaway, then do the exercise. Tool-use focused — use AI you already have.

• SNK3-238toolsAPI key hygiene — the three rulesKeys never go in shared prompts or client code, get rotated when exposed, and get scoped to least privilege.L3–L6 · Verifier–Architect (bootstrap — system rebalances) · ~5m
• SNK3-237toolsNo-code intake: form in, summary outChaining a form to an AI summarize step is the smallest real AI application — and teaches the full build loop.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~8m
• SNK3-236toolsPackage one workflow as a custom assistantCustom GPTs / Claude Projects / Gems let you package instructions plus reference files into a reusable assistant — no code required.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~10m
• SNK3-225toolsOne task, two tools, pick a winnerA 10-minute same-task bake-off on your real work beats any review article for choosing tools.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~10m
• SNK3-224toolsFree vs paid — what the upgrade actually buysPaid AI tiers typically buy better models, longer context, and file/tool features — decide by task, not hype.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~5m
• SNK3-221toolsFind the training toggleConsumer AI tools have settings controlling whether your chats train future models; know where yours is set.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~4m
• SNK3-159toolsRun Grok 4.6 once in Build or CursorGrok 4.6 shipped 12 August as a model you can run in Grok Build and Cursor — not only in Copilot. The skill is one owned task in the harness you already use, then inspect before you publish.L3–L6 · Verifier–Architect (bootstrap — system rebalances) · ~5m
• SNK3-158toolsSwitch Copilot to Grok 4.6 on one real fileGrok 4.6 is now a model you can pick inside GitHub Copilot — CLI, IDE, and cloud. The skill is choosing it for one file you already own, then judging the patch yourself before merge.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~5m
• SNK3-157toolsGive a bot the job — then shut the laptopA teammate bot is working when you can leave. The skill is assigning one finished-work job, closing the lid, and inspecting the output later — not watching the demo video all the way through.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~4m
• SNK3-156toolsHand one overnight job to a teammate botGrok Bot is an always-on teammate with its own cloud computer. The skill is one bounded job plus a human approval gate — not watching a demo, and not building an agent platform.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~5m
• SNK3-114toolsDesktop AI sidekicks see your screen — treat capture as a pasteAlways-on desktop AI helpers that read the screen or clipboard are a continuous paste surface. Before you enable capture, decide what must never appear on screen.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~3m
• SNK3-113toolsCRM / sales AI can draft — you still own the customer sendCRM and sales AI can draft emails, next steps, and activity notes in seconds. Treat every outbound as untrusted until you check the contact, facts, and tone — send is a human gate.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~3m
• SNK3-112toolsPDF / reader AI is a skim — verify the one number you will useDoc/PDF AI summaries and extractions are drafts. Before you file, quote, or decide, open the source page and verify the one number or claim you will act on.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-111toolsCode copilot drafts — you still own the review before mergeGitHub Copilot / IDE AI can write the patch in seconds — treat every suggestion as untrusted draft until you understand it and run your checks. Merge is a human gate.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~3m
• SNK3-109toolsDoc AI summary is a skim — check the one claim you will useCopilot (or Gemini) document summarize is a map, not the territory — before you brief someone or act, open the source and verify the one number, name, or decision you will rely on.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-108toolsAI search cites sources — open one before you trust the answerWeb-grounded AI (Copilot Search, Gemini, Perplexity-style tools) can invent calm-sounding facts — treat citations as the real product and open at least one source before you act or share.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-107toolsNotion AI drafts pages — you still own the share linkNotion AI can summarize, rewrite, and fill a page in seconds — treat every AI edit as a draft until you check facts and who the share link reaches.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-106toolsTeams chat AI catch-up is a draft — verify before you actCopilot “summarize what I missed” in Teams chat is a skim aid, not ground truth — check names, decisions, and links before you reply, commit, or escalate.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-105toolsCalendar AI can draft the invite — you still own SendOutlook/Google calendar copilots will propose times and fill attendees from a chat prompt — treat every invite as a draft until you verify who, when, and the body before Send.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-104toolsMeeting-summary AI is a draft — fix names before you shareTeams/Meet/Zoom AI summaries invent attendees and action items — edit the summary before you paste it into Slack, email, or the CRM.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-103toolsEmail AI can draft — you still own the To line and SendUse Outlook/Gmail AI for reply drafts and tone, but verify recipients, attachments, and facts before Send — never auto-send client or HR mail.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-102toolsDeck AI drafts slides — you still own the story and the sendUse Copilot/Gemini in PowerPoint or Slides for outline and layout drafts, then rewrite the narrative yourself and never auto-share client numbers.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~4m
• SNK3-101toolsAsk the sheet AI for a formula — then check one cellWhen Copilot/Gemini in Excel or Sheets writes a formula, treat it as a draft: spot-check one cell with a known answer before you trust the column.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~4m
• SNK3-100toolsPhone AI chat is still work chat — lock the screen and the pasteMobile AI apps (ChatGPT/Claude/Gemini/Copilot on phone) get the same work data as desktop — use the org-approved app, disable lock-screen widget previews for sensitive drafts, and never dictate client names on a train.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-098toolsVoice mode is still a chat log — scrub before you speakDictation and voice-to-AI still create transcripts your org or vendor may store — treat spoken prompts like typed ones: no client names, salaries, passwords, or unreleased numbers.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-097toolsImage AI at work: brand-safe prompt + human publish gateTreat image/voice generators like any workplace tool: clear brief, no client faces or confidential UI, and a human gate before the asset ships.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m
• SNK3-096toolsUse Copilot inside the app — stop paste-exporting work filesWhen your org has Microsoft 365 Copilot or Google Workspace AI, draft and summarize inside Word/Docs/Excel/Sheets so work files stay in the approved tool — not pasted into a consumer chat.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~4m
• SNK3-095toolsSame UI ≠ same model — check the product model idWhen a product swaps the model behind the same chat/agent UI, re-run your smoke checks — yesterday's quality and cost no longer apply.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~3m
• SNK3-094toolsKeep AI out of your real browser profileAI-enhanced browsers inherit your logged-in sessions — prefer a separate browser profile for AI browsing tools so passwords and work tabs are not the tool surface.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~3m
• SNK3-093toolsAI inside Docs/Sheets is still a send riskWorkspace add-ons put models where you already work — powerful, and easy to leak client data if share/send is still human-owned only in theory.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~4m
• SNK3-090toolsChat → publish is not a free passDescribe-and-deploy AI tools make shipping a link trivial — still run secrets, domain, and human-gate checks before anything is public.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~4m
• SNK3-083toolsConstrain tools before a long AI sessionBefore a long chat/agent session, set tool limits (domain allowlists, disable image/video gen, date cutoffs) so capability is not permission.L2–L5 · Capable–Builder (bootstrap — system rebalances) · ~4m
• SNK3-079toolsLive screen context vs pasted screenshotsScreen share gives continuous visual context — powerful for coaching, wasteful if a 1-line paste would do.L1–L3 · Hands On–Verifier (bootstrap — system rebalances) · ~4m
• SNK3-073toolsYouTube snack: Claude Cowork mobile — land at 0:47Mobile Cowork/chat is for portable assist — still not a free pass for work secrets on a phone.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~5m
• SNK3-071toolsX snack: Grok screen share — AI that sees your work in real timeScreen-share AI can see your work live — powerful for coaching, dangerous if secrets are on screen.L1–L2 · Hands On–Capable (bootstrap — system rebalances) · ~4m
• SNK3-068toolsX-style: NotebookLM meets the feedBring feed ideas into NotebookLM (or grounded tools) so answers cite sources, not hype.L0–L2 · Curious–Capable (bootstrap — system rebalances) · ~3m

Take Question 1 · Team training