Three picks, every week
Skill of the Week, Prompt of the Week, Tool Drop. Plus a monthly stack post — "what I'd use today." Short reads, hand-picked from the library.
- Tool DropJune 16, 2026
Best AI for your role — 7 new pick lists
Three opinionated picks for sales reps, marketers, founders, engineers, CS, cold email, and image generation.
Read - Tool DropJune 16, 2026
Best AI for your role — 9 more pick lists
Designers, HR, PMs, content creators, meeting notes, agencies, consultants, e-commerce, and research synthesis. Three fact-checked picks each.
Read - Prompt of the WeekJune 16, 2026
7 new prompts — research, newsletter, launch, investor emails
Structured paper summaries, cross-study comparison, newsletter issue builder, landing page copy, and two investor update templates.
Read - Tool DropJune 16, 2026
SpaceX acquires Cursor for $60B
SpaceX agreed to buy Anysphere (Cursor) for $60 billion in stock — four days after the IPO, and after all 11 xAI co-founders left. The deal closes Q3 2026.
Read - Tool DropJune 9, 2026
Prompt Engineering gets its own section
Chain-of-thought, prompt chaining, role assignment, structured output — four new technique guides in a dedicated Prompt Engineering section of /learn.
Read - Skill of the WeekJune 5, 2026
Ship an AI Feature on Claude — the gap between a demo and production
A Claude-powered feature is easy to demo and hard to ship. The demo works once on a happy-path input. Production brings weird inputs, costs that creep, hallucinations users see, and no way to tell when quality drifts. This bundle closes that gap.
Read - Skill of the WeekJune 4, 2026
Build, Ship & Host — the workflow bundle for getting a product all the way live
Getting an AI tool to build something is the easy part. Turning it into a product that's live, with sign-in, payments, and a host you won't flee in six months, is where people stall. This bundle is the sequence.
Read - Skill of the WeekJune 3, 2026
Claude API Builder — the three things most Anthropic SDK code skips
Most Claude API code leaves performance and money on the table for the same three reasons every time: no prompt caching, no streaming, and a model version that's already out of date. This skill writes them in from the start.
Read - Prompt of the WeekMay 31, 2026
The file your AI reads before your code
AGENTS.md (or CLAUDE.md, or .cursorrules) is the standing instruction set the AI follows on every turn. Most repos skip it entirely or paste a generic template. Seven sections in order of impact.
Read - What I'd use todayMay 27, 2026
Leaving the builder — four new migration guides under ship-own-stack
The most common question from people who ship something in Lovable, Replit, Emergent or Base44 is "how do I get out of this when I need to?" The ship-own-stack pillar now answers that, per builder, with verified source citations.
Read - What I'd use todayMay 26, 2026
May 2026 stack — understand before you build
This month's shape — a Foundations pillar that explains how AI works under the hood, paired with role-shaped paths through the library. The bet is that operators who understand the mechanism make better choices than ones who just pattern-match.
Read - Tool DropMay 26, 2026
OpusClip — the long-video to short-clip pipeline that stopped being a hack
A podcast, webinar, or YouTube video goes in. Ranked vertical clips with captions and active-speaker framing come out. The category was held together with chewing gum two years ago. It isn't now.
Read - Skill of the WeekMay 26, 2026
LLM Cost + Quality Tuner — the skill that pays for itself in the first hour
Most production LLM features can be 30–70% cheaper without measurable quality loss. The first step is profiling where the money goes. Most teams have not done it.
Read - Prompt of the WeekMay 26, 2026
Plan the change before the agent touches a file
The most common AI-agent failure isn't bad code — it's editing the wrong scope. This prompt forces a file-by-file plan and an explicit out-of-scope section before any edits happen.
Read - Skill of the WeekMay 24, 2026
AI Output Verifier — catch hallucinated APIs before they ship
Today's post lays out the 12 things to check on an AI-generated PR. This skill automates the part that catches hallucinations.
Read - Tool DropMay 18, 2026
Aider — the boring AI pair programmer that just works
No IDE, no extension, no integration hell. A terminal tool that edits files, makes proper git commits, and works with whatever model you bring.
Read - Prompt of the WeekMay 18, 2026
Docs from working code (without the marketing words)
Most AI-generated docs read like product brochures. This prompt forces the output to be the docs you'd actually want to read at 2am while debugging.
Read - Skill of the WeekMay 18, 2026
A code review skill that catches what you'd rationalize away
The Code Review Helper in the Developer pack is sharper than asking the same model that wrote the code. It looks for the bugs you'd skim past at 10pm.
Read - Tool DropMay 4, 2026
Cline 3.0 — MCP-native coding agent
Native MCP support is the headline. The actual story is that Cline is now the most extensible coding agent for solo devs willing to wire things up themselves.
Read - Prompt of the WeekMay 4, 2026
The Three Drafts pattern
Stops AI from converging on the most predictable answer. The single highest-leverage prompt pattern we've found for any kind of writing.
Read - Skill of the WeekMay 4, 2026
Cold outreach that doesn't sound like AI
The Sales pack's cold-outreach skill is the one most people would benefit from running first. Here's how to use it without the usual AI tells.
Read