Your AI resource library
A growing library of guides, prompts and skills for using AI well, no code needed. New here? Start with the map below.
Claude 101
The six building blocks for working with Claude: Projects, Skills and Artifacts to set up, then Connectors, Scheduled tasks and Dispatch to put it to work. New here? Start here.
Open → BeginnerAI Fluency: the foundations
The 4D Framework: how to hand work to AI and still own the result. A short lesson, with the depth in downloadable field guides.
Open → BeginnerHow to Prompt
The basics of asking: set the stage, name the task, show what good looks like. Then how to steer when the first answer misses.
Open → BeginnerUse Cases
Real use cases from Anthropic's own library, curated for business and grouped by kind of work. What people actually do with Claude.
Open → AdvancedThe SOP Generator
Messy notes in, a clean SOP out. Copy-paste prompts that write up how you do your work, then improve it or turn it into a video. The move Sean used to build his own teaching resources.
Open → BeginnerChat vs Cowork: what's the difference?
The question that comes up most. Both are powerful, so what is the difference? You steer Chat, you delegate to Cowork.
Open → AdvancedAdvanced skills
The thinking and building skills I lean on daily: deep-thinking, red-team, prd, build. Teach AI a job once, run it any time with one command. Free.
Open → BeginnerFree tools + starter files
Small finished apps you set up in minutes, like a live price tracker, plus starter files to drop into your own work. A tool, not a command.
Open → BeginnerThe Leverage Ladder
The whole map as one picture: how AI gets more useful, from a single prompt up to helpers that do a whole job for you.
See the ladder → AdvancedThe 4D Build
A skill you run to actually ship something with AI. Walk one idea through the four Ds until it is real, then keep the result.
Open → AdvancedThe Translator
Describe your job in plain words. It interviews you, names what good looks like, and writes the prompt you could not.
Open → AdvancedSubagents
Delegate to Claude the way you delegate to people. What the context window is, when to hand work to a helper, and the brief that comes back done.
Open → AdvancedGood loops + the receipts
Make AI check its own work and get genuinely better, not just done. The studies that back it up now live here too, sources included.
Open → AdvancedMission Control
The top of the ladder, made visible: one mission fanned out to a team of agents, checks landing live. The four Ds, as instruments.
Open → BeginnerFAQ
The questions that come up in nearly every lesson, answered once: API keys, usage limits, data safety, Pro vs Max, and more.
Open → BeginnerVocab gym
Flip a card to check the terms you have learned actually stuck. Two minutes, surprisingly fun.
Have a go →Worth your time
A short, vetted shortlist beats the firehose. Start at the top, stop when you have enough.
More stars mean start here. ★★★start, ★★go deeper, ★browse.
Straight from the people who make Claude: what is new, what each surface does, and how to use it well. Trust this over any third party for Claude facts.
The official documentation. Dry but reliable when you need the real answer on a feature, a limit, or how something actually works.
Clear walkthroughs of AI automation and agent workflows for people who are not engineers. Good for seeing a whole build end to end.
Talks, feature demos, and research explainers from Anthropic itself. Slower-moving than the news channels but authoritative.
Want our own worked-through findings rather than other people’s videos? The receipts on the loops page hold the studies we have checked ourselves, with the sources.
New prompts, loops and guides land here often, so check back. Want it set up around your own work? Book a free chat and we can do it together: calendly.com/seanmccloskey/30min.
McCloskey.ai · Resources. Built for people who want to use AI well without writing code.