AI Tools & Skills
A working guide to the AI tools, models and skills I use for agentic coding. Read it top to bottom: set up your accounts and machine, meet the coding agents, then give them context, skills and loops.
Getting started
Start here: the accounts and the machine you set up before anything else.
Accounts
A handful of sign-ups unlock everything below — most are free to try, and the paid ones bill either monthly or by the token.
GitHub
Free
Where code and Agent Skills live. Make this account first — you will clone repositories, keep your own work, and install most of the skills on this page from here.
Anthropic
Claude Pro from $20/mo · API pay-as-you-go
The maker of Claude and Claude Code. A Pro or Max subscription powers the coding agent; an API key covers anything billed by the token.
OpenAI
ChatGPT Plus from $20/mo · API pay-as-you-go
The maker of ChatGPT and Codex. The same ChatGPT plan that runs the chat app also drives the Codex coding agent.
Google AI Studio
Free tier
A free Gemini API key in a couple of clicks. Handy for image work and quick model tests.
Figma
Free tier
Design files, and the account behind Figma Make and the Figma MCP used further down this page.
Hardware
You need a real machine to drive these agents — a phone or iPad will not do.
Mac mini
From $599
The cheapest way in, and a natural always-on box to leave on your desk. Spec the memory up rather than the looks.
MacBook Air
From $999
Possibly the best deal in tech right now, and plenty for driving agents on the move. It’s fanless, though, so it throttles under the sustained heat of running models locally — for that, you want the Pro.
MacBook Pro
From $1,599
The one to buy if you genuinely want to run models on-device. The fan is the point — it cools the chip through sustained inference, where the Air would throttle. Pair an M-series Pro chip with 64GB+ and a chunk of the work costs nothing per token.
Coding agents
The agents that carry a task from request to a shipped change.
Coding agent
Claude Code
Anthropic’s agentic coding tool — terminal, IDE or web, carrying a task from request to a shipped change. It’s the stronger coder for me. One catch: the subscription only powers Claude Code itself. To drive other agents you drop down to the pay-per-token API, which adds up fast.
Pro $20/mo · Max $100 or $200/mo — for real work you’ll want at least Max
Coding agent
Codex
OpenAI’s coding agent for terminal, IDE and web. Its quiet advantage: the same ChatGPT plan can power other agents too — Hermes can run on your Codex subscription, where Claude Code would bill you separately through the API. On sharing a plan, Codex wins.
Plus $20/mo · Pro $100 or $200/mo — the $100 Pro tier is enough to run agents on your plan
Command-line tools
Where the agents actually run — the terminal, and the tools that keep a fleet of them in view.
Terminal
cmux
A native macOS terminal built on Ghostty, with vertical tabs, split panes and notification rings — built for running several AI coding agents side by side.
Open source · Free
cmux setup guide
Running agents in parallel
Artem Zhutov’s guide to driving a fleet of agents from cmux — an orchestrator on top, with a dashboard, sessions and workspaces underneath, so ten tabs stop becoming ten guesses.
Free
Menu-bar tools
Small native apps that keep an agent fleet in view.
CodexBar
Open source · Free
Usage limits, credit balances and reset countdowns for Codex, Claude Code and 40+ AI coding providers, all from the macOS menu bar.
RepoBar
Open source · Free
Keeps GitHub work in view from the menu bar — CI status, open issues, pull requests, releases and rate-limit health for every repo you follow.
Knowledge & context
How I feed and maintain the context the agents work from.
Knowledge base
Obsidian
A local-first markdown knowledge base. Your notes stay as plain files on your own machine, linked into a graph you control — the store everything else feeds.
Free for personal use
Plugins & knowledge bases
The add-ons I run on top of the vault, plus a knowledge base worth pointing agents at.
Web Clipper
Free
A browser extension that saves any web page as clean markdown straight into your vault, with templates for the metadata you want.
Sync
$8/mo
I run the $8 Sync plan to keep my vaults in step — from the server down to the Mac, and across every device — end-to-end encrypted, with version history I can roll back.
Dataview
Open source
Turns your notes into a queryable database — list, table and task views generated live from frontmatter and inline fields.
Excalidraw
Open source
Hand-drawn diagrams and sketches you draw straight inside a note, kept alongside the writing.
Bases
Free with Obsidian
Obsidian’s built-in way to turn a folder of notes into a database view — filter, sort and group without leaving the vault.
Graph
Free with Obsidian
The graph view — every note a node, every link an edge, so the shape of what you know is something you can see.
LLM wiki
Free
His pattern for AI-maintained knowledge bases: the model builds a cross-linked markdown wiki that compounds as sources and questions accumulate.
Skills & commands
Everything that extends an agent: the skills it runs, the commands that drive it, and where to find more.
Anthropic skills
The Anthropic skills I reach for most — document work and builders.
Open source
Fills forms, merges and splits files, and pulls text and tables out of PDFs — the document work that used to mean a dedicated tool.
docx
Open source
Creates and edits Word documents with tracked changes, comments and formatting kept intact.
xlsx
Open source
Builds and edits Excel workbooks — formulas, formatting and multiple sheets — and reads the numbers back.
pptx
Open source
Generates and edits PowerPoint decks, keeping layouts, themes and speaker notes in place.
skill-creator
Open source
Authors new skills — scaffolding the folder, instructions and examples that teach Claude a repeatable task.
mcp-builder
Open source
Generates MCP servers, wiring a tool or API into something an agent can call directly.
Claude Code commands
The built-in commands, and what I use each one for.
/plan
Built in
Researches the codebase and lays out an approach before changing anything. You approve the plan, then it executes. Toggle it with Shift+Tab.
/goal
Built in
Set a finish line and it keeps working across turns, checking after each one, until the condition is met.
/workflows
Built in
Watch and manage a workflow — a script that runs many subagents in parallel — as it works in the background.
/loop
Built in
Runs a prompt on a repeating interval, or paces its own iterations when you leave the timing to it.
/agents
Built in
Defines subagents you can hand specialised work to, each with its own tools and instructions.
/rewind
Built in
Rolls the code and the conversation back to an earlier checkpoint when a path turns out wrong.
Claude Code plugin
Compound engineering
Kieran Klaassen's plugin for Claude Code, built on a single idea from Every: every piece of work should leave the next one easier. Solved problems get written down as you go, so the setup compounds the more you use it. It ships 26 agents, 23 commands and 13 skills.
Open source · Free
The loop
Brainstorm → Plan → Work → Review → Compound, then repeat. The fourth step is the one most AI coding setups skip.
Brainstorm
Turns a fuzzy idea into a sharp, agreed requirement before any code is written.
Plan
Researches the codebase and writes an implementation blueprint to work from.
Work
Executes the plan in an isolated worktree, validating as it goes.
Review
Runs a panel of specialist reviewers in parallel — security, performance, architecture and more.
Compound
Writes the solved problem into docs/solutions/, so the next run starts ahead.
LFG
The end-to-end pipeline: one command takes an idea all the way to a pull request.
Reviewers and skills
Behind the loop sit 14 specialist review agents — security, performance, architecture, data integrity, code quality and framework-specific reviewers for Rails, Python and TypeScript — plus 13 skills holding the architecture patterns, style guides and design systems the agents read while they work.
Notable skills
Individual skills worth borrowing.
retrospective
Free
Artem Zhutov’s /retro. At the end of a session the agent looks back, works out what it learned and writes it down for next time. Skip it and every session starts from zero; run it and the learning compounds across everything you do.
handoff
Free
Artem Zhutov’s /hand-off. A context window fills up and the agent dulls in the last fifth. /hand-off packages the live state — the work, the session, where you are — and passes it clean to a fresh agent, so a long task survives the boundary.
skill-management
Free
Artem Zhutov’s approach to growing a skill. Start with everything in one SKILL.md. When that file gets too big, split the parts into workflows and let the main file route to them — so the agent loads only the workflow it needs.
grill-me (Matt Pocock)
Open source
From Matt Pocock’s engineering skills. The grill-me skill interviews you about a plan until every branch of the decision tree is resolved.
gstack
Open source
Garry Tan’s opinionated Claude Code setup — 23 tools that act as CEO, designer, engineering manager, release manager and QA inside your agent.
Where skills come from
Where Agent Skills come from.
GitHub
Free
Most open Agent Skills ship as a GitHub repo — a folder of instructions, scripts and examples you can read, fork and drop into your own agent.
ClawHub
Free
A skill marketplace for Claude Code and OpenClaw with vector search — the closest thing yet to npm for AI agents.
Peter Steinberger
Free · GitHub
Ex-PSPDFKit founder, now at OpenAI and steward of OpenClaw. His GitHub is the source behind CodexBar and ClawHub — two of the tools already on this page — plus agent-scripts, a shared rule set for coding with agents.
Loops & hosted agents
Agents that run on their own — on a loop, or on a server that never sleeps.
Loops & autonomy
Point an agent at a metric or a schedule and it keeps going — optimising, repeating, iterating — until the job is done or you call it off.
Loop Library
Free
Forward Future's collection of practical agent loops — reusable prompts with clear success checks and stopping conditions, across engineering, evaluation, operations, content and design. Copy a loop straight in, or install the Loopy skill to run them from your agent.
/loop
Built in
Run a prompt or a command on a recurring interval — a quick way to poll, watch or repeat a task without minding it.
looper
Free
Design and scaffold an agent loop: set the scope, the metric and the keep-or-revert rule, and it writes the harness for you.
autoloop
Free
The narrow optimiser. Edit, measure, keep the win or revert, and repeat overnight against a single number you set.
autoresearch
Open source
His loop for autonomous ML research: the agent edits training code, runs short timed experiments and keeps only the changes that move the metric.
Hosted agents
The always-on side: an autonomous agent on a rented server, working while the Macs are asleep.
Hermes (Nous Research)
Open source
Nous Research’s autonomous agent. I run it for vault maintenance — filing, organising and keeping my Obsidian notes in order on a schedule.
Hosted server
~€15/mo · 8GB RAM, 80GB
Where both agents live. You need your own server so they’re always on — I don’t keep a computer running and online around the clock, so I rent a Hetzner box. Aim for the 8GB RAM, 80GB tier — enough headroom for the agents and their state.
Tools & data
The specialist tools I reach for, and the data and directories behind the investing work.
Design & image generation
Making images, mocking up designs, and styling the output.
ChatGPT Images 2.0
Image gen in ChatGPT · Plus from $20/mo
My favourite image generator now. It follows a detailed brief closely and lets you edit a picture conversationally until it’s right.
Google AI Studio
Free tier
Google’s playground for Gemini, including the Nano Banana image model — strong at editing a picture and holding a subject steady across generations.
Figma Make
Free tier · Pro from $16/seat/mo (AI credits)
Figma’s AI generation tool. Describe an interface and it builds a working prototype you can edit by hand — the part of Figma I keep coming back to.
Frontend Slides
Open source · Free
A Claude Code plugin that builds animation-rich HTML slide decks from a prompt or an existing PowerPoint. Zero dependencies, a library of bold templates and a design philosophy aimed squarely at avoiding AI slop.
Tufte CSS
Open source · Free
Tools for styling web articles after Edward Tufte’s books — generous sidenotes, tight text-and-figure integration and careful typography.
Tufte data viz (Caylent)
Open source
A Claude Code skill that applies Edward Tufte’s principles to generate clean charts across Recharts, ECharts, Chart.js, matplotlib, Plotly and D3.
Figma use (MCP)
Free tier · Pro from $12/seat/mo
Figma’s MCP server brings designs into the workflow — an agent can read a file’s structure, variables and components, and turn a frame into code.
Local tools & utilities
Smaller tools I reach for around the agents.
Ollama
Open source · Free
Runs small language models on your own machine. With enough memory you keep routine work local and off the token meter.
Playwright
Open source · Free
Drives a real browser from code — clicking, typing, screenshotting. It’s how the agents capture pages and check their own work.
Marker
Open source
Turns PDFs, slides and documents into clean markdown the agents can read — tables, structure and all.
birdclaw
Open source
Steipete’s local-first tool for your X account. It pulls your tweets, DMs, likes and bookmarks into a database you own, with a web view to search, read and post.
Typora
$14.99 one-time · 15-day trial
Where Obsidian holds the linked vault, Typora is for the single document. A clean, distraction-free live preview for drafting one piece and exporting it to PDF, Word or HTML.
FluidVoice
Open source · Free
A free, open-source Mac dictation app. Its on-device Fluid-1 model cleans spoken words into properly punctuated text in any app, across 40+ languages, with nothing leaving the machine.
Model comparison
Where I check which model to reach for.
Artificial Analysis
Free
Independent benchmarks that compare models and providers across intelligence, speed and price, so you can pick the right one for a job.
Arena (LMArena)
Free
A crowd-sourced leaderboard that ranks models by blind, side-by-side human votes — millions of them, scored with Elo ratings.
Market & financial data
The data feeds behind the investing work.
Yahoo Finance
Free
Free market data — quotes, history and fundamentals — easy to pull for a quick look or a chart.
FRED
Free
The St. Louis Fed’s economic database: hundreds of thousands of macro series, free, and my first stop for rates, inflation and growth.
Financial Modeling Prep
Free tier
Real-time quotes and 30+ years of fundamentals over a clean API. The free tier covers a lot before you ever pay.
fiscal.ai
Free tier · paid plans
Clean, auditable financial statements updated within minutes of earnings — the deep-dive data behind the bvest analysis.
DefiLlama
Free
Open DeFi data — total value locked, fees and volumes across chains and protocols. The crypto-side data source.
Directories & lists
Curated lists for finding evals, MCP servers and connectors.
Awesome Evals
Free · GitHub
A curated library of the best resources for building and evaluating AI agents — papers, talks, tools and benchmarks, plus deep reading notes on eval frameworks like Inspect AI and promptfoo, LLM-as-judge and agent-trajectory testing.
Awesome MCP Servers
Free · GitHub
A large community list of MCP servers — the connectors that hand agents real tools. The travel and transportation section alone runs from live flights and hotels to campground search, transit schedules and route data.
Awesome Claude Connectors
Free · GitHub
A directory of Claude's connectors — hundreds of verified MCP integrations sorted into 30+ categories with descriptions and use cases. The travel section covers Kayak for flights, hotels and cars and Tripadvisor for reviews.
People & my work
Who I learn from, and the skills I have built for myself.
People I follow on X
The accounts I learn the most from.
My skill families
Families of skills I have built for my own workflows.
bvest
Coming soon
An investing workflow built as a hedge-fund org chart — macro strategist, analysts, an investment committee and a portfolio manager, each its own seat.
bknow
Coming soon
Knowledge-management skills for Obsidian vaults: filing, frontmatter, backlinks and synthesis across notes.
blook
Coming soon
A design-output engine. It routes brand tokens into charts and tear sheets, then exports to HTML and drives PowerPoint and Word through scripts — everything coming out on-brand.
bvoice
Coming soon
A writing pipeline — storyboard, draft, edit and rewrite — that holds a consistent voice from notes to published piece.
bdas
Coming soon
Deal and strategy skills covering the M&A lifecycle, from thesis and fieldwork through to the final report.
brun
Coming soon
Runtime operations for the agent fleet — health checks, cron, skills and repair across the machines.