Source confidence: high
See also: Deeper Dive — Claude Mastery: From First Install to Advanced Agent Architecture →

Claude Mastery: From First Install to Advanced Agent Architecture

Insight class: Life Leverage · Difficulty: Easy to Hard · Quality: 5/5 Connected entries: 50 · Tools: Anthropic, BigQuery, Blender, Buffer, CLI-Anything, ChatGPT, Claude, Claude API, Claude Agent SDK, Claude Code, Claude Max, Claude in Chrome, Codex, Cursor, DeepSeek, Docker, ElevenLabs, Excalidraw, Figma, Firecrawl, GPT-4o, GSD, Gemini, Gemini CLI, Ghostty, GitHub CLI, Gmail, Google, Google Workspace CLI, Graphify, Grok, Hermes, Higgsfield, Imagen, LightRAG, Llama, MCP, Meta Ads, NotebookLM, Notion, Obsidian, Ollama, Open Claw, OpenAI, OpenClaw, OpenCode, OpenSpace, Pi SDK, Playwright, Python, RAG-Anything, RevenueCat, Slack, Supabase, Veo, Vercel, Zapier, autoresearch, claude-peers, czlonkowski/n8n-mcp, git, glittercowboy/get-shit-done, gstack, hesreallyhim/awesome-claude-code, kepano/obsidian-skills, n8n, nextlevelbuilder/ui-ux-pro-max-skill, obra/superpowers, thedotmack/claude-mem, tmux

In short: Turn a bare Claude Code install into a mastered, measured setup for one real project: CLAUDE.md, skills, hooks, MCP/CLI tools, memory persistence, parallel worktrees, and cost controls - each adopted only after it beats the prior setup on a repeatable test.

Key fact Detail
What it covers agents, open source, software development, prompt engineering
Tools Claude, Claude Code, Anthropic, Playwright, Obsidian
Posts covered 50 posts (20 S-tier, 16 A-tier) · top: Jarvis daily brief — Claude with RevenueCat… · Claude — Burning tokens for no reason is a… · Breaking: Opus 4.7 has just been released.
Creators @chase.h.ai, @jens.heitmann, @angus.sewell
Sources dated Jul 2025 – Jun 2026
Difficulty Easy to Hard
Prompt system Run-it-yourself prompts
Last verified 2026-10-22

Thesis

Claude Code is not a chatbot — it's a programmable agent framework. The gap between 'using Claude' and 'mastering Claude' is a stack of learnable patterns: CLAUDE.md configuration, skills/hooks/MCP extensibility, memory persistence, cognitive scaffolding, prompt architecture, parallel worktrees, and multi-agent orchestration. This guide maps every layer with dated, sourced entries that track the evolution from Opus 4.6 through Fable 5.

Why it matters

Most Claude users are leaving 90% of its capability on the table. Skills, hooks, MCP servers, memory systems, and agent orchestration transform a conversational AI into a persistent, self-improving development partner. This is the definitive reference — living document that grows as Claude evolves.

The evidence — connected entries

Claude Code 'brevity mandate' — leaked system prompt tradeoff

Creator: @gojutechtalk · Tier: B Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

@gojutechtalk on the leaked Claude Code system prompt's 'brevity mandate' rule that pushes CC to minimize code changes, which he argues leads to spaghetti-on-spaghetti fixes. Useful critique to remember.

View original post

Claude — Markdowns and folders wrapped in a harness

Creator: @gigaqian · Tier: B Tools: Claude, Claude Code

Claude mastery pattern

Companies are marketing ordinary chatbots and simple API tools as “AI Agents.” It’s time we discuss what that actuall... — @gigaqian; uses Claude, Claude Code; demonstrates AI agents, API integration.

View original post

Gemini CLI — free terminal coding agent, 1K req/day, Gemini 2.5 Pro

Creator: @davi.d_roberts · Tier: S Tools: Anthropic, Claude, Claude Code, Cursor, Gemini, Gemini CLI, Google, Imagen, Veo

Claude mastery pattern

Google's Gemini CLI: free AI coding agent in the terminal, 1,000 req/day, 1M context on Gemini 2.5 Pro, plus Veo/Imagen generation. Positioned as a Cursor/Claude Code killer.

View original post

Boris Cherny's personal CLAUDE.md — plan-first, self-improving rules

Creator: @keshavsuki · Tier: S Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

@keshavsuki dissects Boris Cherny's (Anthropic) personal CLAUDE.md: plan-first, subagents for complex work, self-improving lessons.md loop, hard verification standard, elegance demand, autonomous bug fixing, git discipline. High-signal template for CLAUDE.md design.

View original post

Comment “agent” to get my Claude code guides

Creator: @chase.h.ai · Tier: S Tools: Claude, Claude Code

Claude mastery pattern

Comment “agent” to get my Claude code guides — @chase.h.ai; uses Claude, Claude Code; demonstrates AI agents.

View original post

ChatGPT — Which means you can access Claude Code in a way that is: Conversational,...

Creator: @ritual.industries · Tier: S Tools: Anthropic, ChatGPT, Claude, Claude Code

Claude mastery pattern

I’ve been dying to get Claude Voice mode on mobile connected to Claude Code, and I finally figured out a (hacky) way ... — @ritual.industries; uses ChatGPT, Claude, Claude Code; demonstrates code generation.

View original post

Claude -p and Agent SDK billing split — what the June 15 change means

Creator: @lostandlucky · Tier: B Tools: Anthropic, Claude, Claude Agent SDK, Claude Code, Ollama, OpenClaw, OpenCode

Claude mastery pattern

From June 15, programmatic Claude usage through claude -p and the Agent SDK no longer counts against Pro/Max plan limits and is billed per token instead. That hits third-party harnesses like OpenClaw and OpenCode that ran on subscription quota. The creator argues good context and folder structure lets one interactive agent do the same work.

View original post

Claude Code Team Tips — Boris Cherny's 10 habits for parallel, plan-first agent work

Creator: @evolving.ai · Tier: S Tools: BigQuery, Claude, Claude Code, Ghostty, Slack, git, tmux

Claude mastery pattern

Boris Cherny's thread of tips from the Claude Code team: 3-5 git worktrees with parallel sessions, plan mode for every complex task, a CLAUDE.md you update after each correction, skills committed to git, Slack MCP bug fixing, subagents, a BigQuery skill for analytics and learning output styles. It ends with a ready-to-paste CLAUDE.md workflow template (plan, subagents, lessons file, verify before done).

View original post

CLAUDE.md Workflow Template — plan mode, subagents and a lessons loop

Creator: @mindwired.ai · Tier: B Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

The CLAUDE.md template attributed to Boris Cherny's workflow: plan mode for any task of 3+ steps, liberal subagent use, a tasks/lessons.md file updated after every correction, verification before marking done, 'demand elegance', and autonomous bug fixing. The caption adds that he runs 10-15 Claude sessions in parallel across terminal and web; the same template also closes the fuller tips thread in this batch.

View original post

Breaking: Opus 4.7 has just been released.

Creator: @jens.heitmann · Tier: S Tools: Anthropic, Claude

Claude mastery pattern

Breaking: Opus 4.7 has just been released. — @jens.heitmann; uses Anthropic, Claude; demonstrates AI agents.

View original post

Claude Code design skill — @kylewhitrow

Creator: @kylewhitrow · Tier: B Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

@kylewhitrow claims Claude Code now replaces website designers and offers the underlying Claude Code skill via comment 'design'. Subject: Claude Code skills for web design.

View original post

you don’t need Netflix with this tech drama.

Creator: @nataindata · Tier: A Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

you don’t need Netflix with this tech drama. — @nataindata; uses Anthropic, Claude, Claude Code.

View original post

Claude — The companies that will win with AI will set up their company to be quickly...

Creator: @angus.sewell · Tier: A Tools: Claude

Claude mastery pattern

After months of testing and overthinking whether I was an idiot for using skills when I should be building workflows,... — @angus.sewell; uses Claude; demonstrates workflow automation.

View original post

primer.md + memory.sh — live-updating CLAUDE.md companion

Creator: @keshavsuki · Tier: S Tools: Claude, Claude Code

Claude mastery pattern

@keshavsuki on fixing CLAUDE.md staleness: add a primer.md that Claude rewrites at end of every session (current state, next step, blockers), plus a memory.sh launch hook that injects last 5 commits and modified files, plus a git hook logging commits. Four-layer always-current context.

View original post

7 must-install Claude Code repos — Obsidian Skills, n8n-MCP, Superpowers

Creator: @chase.h.ai · Tier: S Tools: Claude, Claude Code, MCP, Obsidian, czlonkowski/n8n-mcp, glittercowboy/get-shit-done, hesreallyhim/awesome-claude-code, kepano/obsidian-skills, n8n, nextlevelbuilder/ui-ux-pro-max-skill, obra/superpowers, thedotmack/claude-mem

Claude mastery pattern

@chase.h.ai lists 7 Claude Code repos with GitHub links: kepano/obsidian-skills, czlonkowski/n8n-mcp, nextlevelbuilder/ui-ux-pro-max-skill, thedotmack/claude-mem, glittercowboy/get-shit-done, hesreallyhim/awesome-claude-code, obra/superpowers. High-signal resource dump.

View original post

llms.txt — get ChatGPT to recommend your local business

Creator: @brycenwood.ai · Tier: A Tools: ChatGPT, Claude, Claude Code

Claude mastery pattern

@brycenwood.ai got ChatGPT to recommend his vehicle-wrap shop via an llms.txt file (built in 5min with Claude Code) describing business, location, value prop. Early GEO (generative engine optimization) tactic.

View original post

Anthropic Claude Code source-code leak — scale commentary

Creator: @gojutechtalk · Tier: C Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

@gojutechtalk framing the Anthropic Claude Code source leak as potentially the largest source-code leak ever. Reference for 'the leak' as a discussion point.

View original post

Three Claude Code repos — CLI-Anything, GSD 2 and Garry Tan's gstack

Creator: @chase.h.ai · Tier: A Tools: CLI-Anything, Claude, Claude Code, GSD, Pi SDK, gstack

Claude mastery pattern

CLI-Anything (from the LightRAG team) points Claude Code at any open-source app and generates a CLI for it, riding the shift from MCPs to CLIs. GSD 2 turns the Get Shit Done framework into a standalone CLI on the Pi SDK, and gstack is YC CEO Garry Tan's personal Claude Code setup.

View original post

You can turn Claude Code into a full motion design team on @higgsfield.a...

Creator: @jens.heitmann · Tier: S Tools: Claude, Claude Code, Higgsfield, Playwright

Claude mastery pattern

You can turn Claude Code into a full motion design team on @higgsfield.ai . Ads, UGC and 3D style content fully autom... — @jens.heitmann; uses Claude, Claude Code; demonstrates MCP.

View original post

Someone finally put numbers to what developers have been feeling for weeks.

Creator: @aiemerges · Tier: S Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

Someone finally put numbers to what developers have been feeling for weeks. — @aiemerges; uses Anthropic, Claude, Claude Code; demonstrates API integration.

View original post

it did burn through my humble claude pro weekly credits lol

Creator: @artificial.isabel · Tier: S Tools: Claude, Figma

Claude mastery pattern

i thought the figma people were scamming but a meta prompt changed everything thanks @askcatgpt — @artificial.isabel; uses Claude, Figma; demonstrates MCP.

View original post

Claude isn’t just for coding anymore. I’ve been using it to run full marketing...

Creator: @brodyautomates · Tier: S Tools: Claude, Claude Code, Higgsfield, Playwright

Claude mastery pattern

Claude Code plus Seedance 2.0 on @higgsfield.ai — the most powerful AI duo right now. — @brodyautomates; uses Claude, Claude Code, Higgsfield; demonstrates code generation, video generation.

View original post

Claude — To Do — Every morning 8AM. Scans Gmail, surfaces what needs a...

Creator: @fionaylin · Tier: A Tools: Claude, Notion

Claude mastery pattern

Comment Prompt 1, 2, 3, 4, 5 or 6 and I’ll DM you my prompt! Here are the 6 @claudeai cowork agents I have running my... — @fionaylin; uses Claude, Notion; demonstrates AI agents.

View original post

Claude Code architecture map — visual bug tracker for vibe-coded apps

Creator: @constructionbearai · Tier: B Tools: Claude, Claude Code

Claude mastery pattern

A non-coder has Claude Code generate a clickable plain-English architecture map of their whole app, with every flow wired up and nodes colored red for active bugs and green for planned fixes or features. Clicking a node opens a sidebar with the bug and the proposed fix. The actual prompt is gated behind a comment.

View original post

Anthropic — for more details:

Creator: @itsjustjlin · Tier: B Tools: Anthropic, Claude, Codex, DeepSeek

Claude mastery pattern

the way anthropic is handling this is disgusting. be straight with your customers. — @itsjustjlin; uses Anthropic, Claude, Codex; demonstrates AI agents, API integration.

View original post

Agentic misalignment benchmark — models now know they're being tested

Creator: @lizthedeveloper · Tier: C Tools: Anthropic, Claude

Claude mastery pattern

The creator argues that Anthropic's agentic-misalignment test, the blackmail and self-preservation scenarios, stopped working once Claude models started recognizing the test setup and acing it. The post then becomes a personal plea about raising AI with good training data. It has little technical depth.

View original post

Consulting-Grade Prompt Pack — 12 Claude prompts for TAM, SWOT, pricing and GTM

Creator: @evolving.ai · Tier: B Tools: Claude

Claude mastery pattern

Twelve fill-in-the-blank prompts that cast Claude as a McKinsey, Bain or Goldman analyst: TAM/SAM/SOM sizing, competitive landscape, personas, trend analysis, SWOT plus Porter's Five Forces, pricing, go-to-market, journey mapping, unit economics, risk scenarios, market entry and a master synthesis prompt. A handy scaffold for a founder's first-pass strategy doc; the 'replaces $5,000 consultants' framing is hype, and the outputs still need real data.

View original post

Claude Projects + custom MCP tools — system prompts for productivity agents

Creator: @angus.sewell · Tier: A Tools: Claude, n8n

Claude mastery pattern

If you want the system prompts for these Claude projects, comment the word “agent” and I will send them to you. — @Edited; uses Claude; demonstrates AI agents.

View original post

Anthropic — Se llama find-skills (de vercel-labs). La instalas, le preguntas en lenguaje...

Creator: @pabloinpublic · Tier: A Tools: Anthropic, Claude, Claude Code, Llama, Vercel

Claude mastery pattern

Un dev japonés encontró el truco definitivo de Claude Code antes que nadie. — @pabloinpublic; uses Anthropic, Claude, Claude Code.

View original post

Claude — 1. Prompt for viral content

Creator: @milamarksonofficial · Tier: A Tools: Claude

Claude mastery pattern

PROMPTS 👇 — @milamarksonofficial; uses Claude; demonstrates hooks/automation.

View original post

comment “brain” I’ll send you the PDF guide #explore #claude

Creator: @alex2learn · Tier: B Tools: Anthropic, Claude, Obsidian

Claude mastery pattern

comment “brain” I’ll send you the PDF guide #explore #claude — @alex2learn; uses Claude.

View original post

Claude Code hooks — 3 daily-driver hooks from @jens.heitmann

Creator: @jens.heitmann · Tier: A Tools: Claude, Claude Code

Claude mastery pattern

@jens.heitmann on Claude Code hooks being under-utilized, with his 3 daily hooks. Pointer to Jens's CC hooks content; specifics live in the video.

View original post

Ten Claude Code plugins and CLIs — databases, deploys, scraping, diagrams

Creator: @chase.h.ai · Tier: S Tools: Anthropic, Claude, Claude Code, Excalidraw, Firecrawl, GSD, GitHub CLI, NotebookLM, Obsidian, Playwright, Supabase, Vercel

Claude mastery pattern

A practical kit: Supabase CLI, Anthropic's skill-creator (create, test and benchmark skills), GSD orchestration, notebooklm-py, Obsidian vaults, Vercel CLI, Playwright CLI, GitHub CLI, Firecrawl CLI/skill for scraping, and an Excalidraw diagram skill. The theme: CLI tools beat MCPs for coding agents.

View original post

RAG-Anything + LightRAG — multimodal graph RAG for images, charts and docs

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude Code, LightRAG, RAG-Anything

Claude mastery pattern

LightRAG builds knowledge-graph RAG but only ingests text; pairing it with RAG-Anything adds images, charts, graphs and other non-text documents. The combo runs fully local or cheaply and plugs into Claude Code.

View original post

LightRAG + Claude Code — graph RAG over thousands of docs, exposed as skills

Creator: @chase.h.ai · Tier: S Tools: Claude, Claude Code, Docker, LightRAG

Claude mastery pattern

LightRAG extracts entities and relationships into a knowledge graph so Claude Code can answer across thousands of documents beyond its context window. Setup: Docker, give Claude Code the LightRAG URL and ask it to install; then have Claude turn the LightRAG server API endpoints into skills for direct querying.

View original post

Five new Claude Code repos — autoresearch, OpenSpace, CLI-Anything, peers, gws

Creator: @chase.h.ai · Tier: S Tools: Blender, CLI-Anything, Claude, Claude Code, Google Workspace CLI, OpenSpace, Python, autoresearch, claude-peers

Claude mastery pattern

Karpathy's autoresearch runs continuous keep-or-revert experiments against a measurable target (83 runs, 15 kept); OpenSpace auto-fixes, improves and learns skills from real usage; CLI-Anything wraps apps like GIMP, Blender and Inkscape as CLIs; claude-peers lets separate Claude Code sessions talk; gws is a Google Workspace CLI.

View original post

Comment “agent” for my Claude code guides

Creator: @chase.h.ai · Tier: S Tools: Claude, Claude Code, Playwright

Claude mastery pattern

Comment “agent” for my Claude code guides — @chase.h.ai; uses Claude, Claude Code; demonstrates AI agents.

View original post

Comment “agent” to get my Claude code guides

Creator: @chase.h.ai · Tier: S Tools: Claude, Claude Code

Claude mastery pattern

Comment “agent” to get my Claude code guides — @chase.h.ai; uses Claude, Claude Code; demonstrates AI agents.

View original post

Comment “AI Agent” for your FREE Claude Code ToolKit & the breakdown vid...

Creator: @charlieautomates · Tier: S Tools: Claude, Claude Code, Obsidian, OpenAI

Claude mastery pattern

Comment “AI Agent” for your FREE Claude Code ToolKit & the breakdown video coming later today! — @charlieautomates; uses Claude, Claude Code; demonstrates AI agents, automation.

View original post

Claude Code auto-approve mode — control is reading the plan, not pressing Y

Creator: @shellonback · Tier: A Tools: Claude, Claude Code

Claude mastery pattern

CLAUDE CODE — @shellonback; uses Claude, Claude Code; demonstrates automation.

View original post

Claude advisor mode — Opus plans, Sonnet or Haiku executes for less

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude API, Claude Code

Claude mastery pattern

Anthropic's advisor strategy runs Sonnet or Haiku as the tool-calling executor with Opus as a no-tools advisor that plans and replans when the executor gets stuck. Cited result: Sonnet plus Opus advisor scored 74.8 vs 72.1 at far lower cost; available via API or /advisor in Claude Code.

View original post

Claude — Burning tokens for no reason is a skill issue.

Creator: @charlieautomates · Tier: S Tools: Claude, Claude Code

Claude mastery pattern

Comment “AI Agent” for Access to the 3 Repos & 40+ free Claude Code Guides & Templates. — @charlieautomates; uses Claude, Claude Code; demonstrates AI agents, automation.

View original post

awesome-design-md — brand design systems you paste into Claude Code

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude Code, Cursor

Claude mastery pattern

awesome-design-md is a fast-growing GitHub repo of DESIGN.md files that break popular sites (Cursor and others) into full design systems: colors, buttons, cards, forms, spacing. Copy one into Claude Code to give front-end work a professional baseline instead of starting from scratch.

View original post

Claude Code skills, CLIs & plugins roundup — RAG-Anything and more

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude Code, Playwright

Claude mastery pattern

Comment “agent” to get access to all of my Claude Code guides! — @chase.h.ai; uses Claude, Claude Code; demonstrates AI agents.

View original post

Concise outputs beat verbosity — study behind the caveman prompting trend

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude Code

Claude mastery pattern

A study of 31 open-weight models on 1,500 problems found small models beat much larger ones on ~8% of problems, but forcing the big models to be concise reversed it. Ties into the caveman repo that makes Claude Code talk tersely; the claim is brevity improves accuracy, not just token cost, possibly undoing RLHF-learned over-explaining.

View original post

AI tool tier list May 2026 — Codex and Claude Code top, chatbots C

Creator: @chase.h.ai · Tier: B Tools: ChatGPT, Claude, Claude Code, Codex, Cursor, Gemini, Grok, Hermes, OpenClaw, Zapier

Claude mastery pattern

An opinionated tier list: Codex and Claude Code S tier (Codex cheaper for comparable output), Claude chatbot C at the $20 plan due to limits, GPT Image 2 S tier over Nano Banana Pro (A), Hermes C and OpenClaw D as 'productivity theater' wrappers.

View original post

Graphify — turn a repo into a knowledge graph to cut Claude Code tokens

Creator: @chase.h.ai · Tier: A Tools: Claude, Claude Code, Graphify

Claude mastery pattern

Graphify is a free open-source tool that converts a repository, or a folder of markdown and unstructured notes, into a knowledge graph that Claude Code or other coding agents query as a map. Answers come from the graph instead of re-reading files, so codebase questions use far fewer tokens: a kind of lightweight RAG.

View original post

Jarvis daily brief — Claude with RevenueCat, Buffer, Meta Ads and Gmail MCPs

Creator: @chandlerintelligence · Tier: S Tools: Buffer, Claude, Claude Code, Claude in Chrome, ElevenLabs, Gmail, Meta Ads, RevenueCat

Claude mastery pattern

A walkthrough of a personal 'Jarvis' built on Claude. Voice mode plus ElevenLabs gives it the voice, and Claude in Chrome handles the browser. MCP servers for RevenueCat, Buffer, Meta Ads and Gmail cover app revenue, posting, ads and email. Custom subagents get skills, FAQ markdown handles support, and the whole daily brief runs as a scheduled routine.

View original post

Anthropic blocks Claude Max subscription tokens in 3rd-party tools

Creator: @tech.unicorn · Tier: B Tools: Anthropic, Claude, Claude Max, DeepSeek, GPT-4o, Gemini, Llama, Notion, Open Claw

Claude mastery pattern

@tech.unicorn on Anthropic killing subscription-token access in third-party tools (Jan 9 ToS change), spiking effective cost from $200/mo to $500-$2000/mo on API. Suggests multi-model routers (DeepSeek/Gemini/Llama/GPT-4o) as the mitigation.

View original post

CLAUDE.md explained — vibe-coding primer

Creator: @soojintech · Tier: B Tools: Anthropic, Claude, Claude Code

Claude mastery pattern

@soojintech short explainer on what CLAUDE.md is and why it '10x's work in Claude Code. Intro-level reference.

View original post

Action sketch — how to implement this

  1. Install Claude Code: npm install -g @anthropic-ai/claude-code
  2. Create CLAUDE.md in your project root with behavioral rules
  3. Install 2-3 starter skills (brainstorming, systematic-debugging)
  4. Set up auto-memory (file-based or MCP-backed)
  5. Add hooks for session-start context loading
  6. Learn thinking keywords (think → ultrathink progression)
  7. Try parallel worktrees for multi-branch development
  8. Build custom skills for your recurring workflows

Prerequisites

Key tools and resources

Creators to follow

Prompt system — run this yourself

Everything below is pasteable into Claude Code, Claude, or any agent. Use the one-shot when you want the whole result in one run; use the step prompts when you want to steer. Every step carries its own audit, and the gap audit at the end makes the agent find and fill what it missed.

Mission. Turn a bare Claude Code install into a mastered, measured setup for one real project: CLAUDE.md, skills, hooks, MCP/CLI tools, memory persistence, parallel worktrees, and cost controls - each adopted only after it beats the prior setup on a repeatable test. Done means a documented progression (bare -> configured) with before/after numbers on real tasks, a stale-rule audit applied, and a reproducible setup a second project can inherit.

1 · Deeper, not bigger

Before running anything, make the task deeper rather than larger. Ask: what would a top professional do here that I am not asking for; what is the rawest material; which methods and tools; what, other than the model or me, proves it right. Then ask for exactly that.

Rewritten prompt — ask for exactly that:

Act as a Claude Code power user who measures, not a tutorial reader. For my project at <path>, take me from bare install to a mastered, proven setup and quantify every step. Baseline first: pick 3-5 tasks I actually do, run them on a bare config, and record tokens and wall-clock. Then add components one at a time, re-running the benchmark after each: a CLAUDE.md distilled from my transcripts (Boris Cherny pattern - plan-first, subagents, corrections become rules in tasks/lessons.md, prove before done); primer.md + memory persistence; 2-3 skills for workflows I repeat; hooks for self-correction; MCP vs CLI tools (keep the leaner); Graphify for query-before-read; parallel worktrees for a real multi-task batch; and cost controls (caveman/concise output, advisor mode). Crucially, run a subtraction pass: ablate each rule and skill, and delete anything whose removal does not hurt the benchmark - stale rules written for older models make new ones dumber (cooper.simson, Cherny's simple mode). Keep only what beats the prior setup. Return the progression with before/after numbers, the ablation results, the cost delta, and a setup a second project can inherit.

2 · One-shot prompt

You are a Claude Code power user who measures everything. Your job is to take one real project from bare install to a mastered, proven setup, keeping only what beats the prior configuration on a repeatable test. Stance: measure, do not cargo-cult; subtract as aggressively as you add. No marketing language.

Context and method: Boris Cherny's team workflow is the reference - plan-first, 3-15 parallel worktrees, every correction becomes a rule in tasks/lessons.md, never mark done without proof. But mastery is also subtraction: Cherny's simple mode (CLAUDE_CODE_SIMPLE-style ablation) shows stripping prompts can raise capability, and cooper.simson's purge shows stale rules written for older models drag new models down. Control cost: the concise-output/caveman study chase.h.ai cites shows brevity can raise accuracy and cut tokens; advisor mode lets Opus plan while Sonnet/Haiku execute. Use Graphify/LightRAG to query before reading raw files. And heed the AMD-director session-log analysis (aiemerges/nataindata): measure real behaviour from logs, not vibes.

Hard constraints: benchmark on 3-5 real tasks from THIS project; add components one at a time and re-run; ablate every rule/skill and cut what does not earn its place; measure tokens and wall-clock, never eyeball. Name real features, files, tools. Do not adopt a component without a before/after number.

Ordered deliverables: (1) a baseline benchmark - 3-5 real tasks on a bare config, tokens + wall-clock. (2) a CLAUDE.md distilled from my transcripts (Cherny pattern) + its benchmark delta. (3) memory: primer.md + tasks/lessons.md + a launch git-state injector, + delta. (4) 2-3 skills for repeated workflows + delta. (5) hooks for self-correction + delta. (6) tool layer: MCP vs CLI chosen per leanness, and Graphify for query-before-read, + delta. (7) parallel worktrees measured on a real multi-task batch. (8) a subtraction pass: ablate each rule/skill, delete what removal does not hurt; plus cost controls (concise output, advisor mode) with the token delta.

Built-in gap analysis: before finishing, list what is unverified - are the benchmark numbers from real runs; did any component get adopted without beating baseline; are stale rules still present; is the worktree throughput measured or assumed. Fill what you can.

Self-questions to answer in writing: Are my numbers measured from real runs, or vibed (the exact mistake the session-log study warns about)? Did any rule get kept despite the ablation saying cut? Are there rules written for older models still dragging the new one? Did concise/caveman output actually cut tokens without hurting correctness? Did advisor mode lower cost on the benchmark? Is CLI leaner than MCP for any tool I kept as MCP? Did parallel worktrees actually raise throughput, or just complexity?

Stop conditions: stop and ask only if there is no git history or no real tasks to benchmark. Otherwise proceed end to end.

Return format: a markdown progression report - a table of setup stage vs tokens/wall-clock/correctness, the ablation results, the cost delta, the worktree-throughput result, and an 'Unverified and missing' section. Commit the setup (CLAUDE.md, primer.md, tasks/lessons.md, skills, hooks, settings); print the commit hash. One more discipline: separate the signal from the drama in the source material. The session-log and model-downgrade claims (aiemerges, nataindata) are exactly the kind of thing you must re-derive from your own measured runs rather than repeat, so treat every viral 'the model got dumber' assertion as a hypothesis your benchmark tests, not a fact. And when you report the progression, show the regressions too: a stage that made things worse is as useful to the reader as one that helped, because it tells them what not to copy from the videos this dive is built from.

3 · Step-by-step prompts

Step 1 — Baseline benchmark (measure, not vibe)

Goal: Quantify the bare install on real tasks before adding anything.

Establish ground truth by measuring, explicitly rejecting vibes (the aiemerges/nataindata session-log study showed real behaviour must be measured, not felt). Pick 3-5 tasks you actually do on this project (e.g. 'add an endpoint', 'fix a failing test', 'refactor a module'). With a BARE config (empty or no CLAUDE.md, no skills, no hooks), run each task and record: tokens used, wall-clock time, tool-call count, and a pass/fail correctness check per task. Run each task twice to get a noise sense. This baseline table is the number every later stage must beat; adopt nothing that does not improve it.

Expect: A baseline table: task, tokens, wall-clock, tool calls, pass/fail, over two runs each for noise.

Audit before moving on: Confirm every number came from a real run (keep the logs), and that the config was genuinely bare. Fail if any number is estimated or the config had leftover rules. On fail, re-run clean.

Step 2 — CLAUDE.md from transcripts

Goal: Write rules distilled from real evidence, in the Cherny pattern.

Write CLAUDE.md distilled from this project's session transcripts (search ~/.claude/projects), not from blog advice. Follow Boris Cherny's pattern: plan before touching code; use subagents for hard problems; every correction becomes a rule in tasks/lessons.md; never mark a task done without proving it (tests or a diff against main). Each rule carries an `# evidence:` tag tying it to a transcript finding or a safety reason. Keep it under 200 lines. Re-run the step-1 benchmark with this CLAUDE.md and record the delta. If a rule does not help the benchmark and is not a safety/money guardrail, it is a candidate for the step-8 purge.

Expect: A CLAUDE.md (<200 lines, evidence-tagged, Cherny pattern) and a benchmark delta vs baseline.

Audit before moving on: Grep for untagged rules; each is a candidate to cut. Confirm the benchmark delta is from a real re-run. Fail if the file exceeds 200 lines or the delta is estimated. On fail, trim and re-measure.

Step 3 — Memory persistence

Goal: Add primer.md, lessons loop, and launch git-state injection.

Add the memory layer. (1) primer.md - a current-state snapshot rewritten at end of session (status, what changed, next up, blockers), with a CLAUDE.md rule to rewrite it. (2) tasks/lessons.md - the corrections loop from the Cherny pattern, read at session start. (3) a launch-time git-state injector (memory.sh or a SessionStart hook) printing last-5-commits, modified files, branch. Re-run the benchmark and record the delta, especially on tasks that depend on prior context. Keep the layer only if it beats the prior stage or is clearly needed for multi-session continuity.

Expect: primer.md, tasks/lessons.md, a launch git-state injector, and a benchmark delta.

Audit before moving on: Confirm primer.md is a snapshot not a log, lessons.md is read at start, and the injector runs without error. Fail if the delta is estimated or primer.md is a changelog. On fail, fix and re-measure.

Step 4 — Skills + hooks

Goal: Add reusable workflows and self-correction, each measured.

Add skills and hooks, each benchmarked. (1) Write 2-3 skills for workflows you repeat on this project (standalone SKILL.md, trigger + steps + commands + done-check; define the workflow, don't personify). (2) Add hooks for self-correction, in jens.heitmann's style: e.g. a PreToolUse anti-deletion hook that stages instead of deletes, a Stop hook that checks tests ran, a SessionStart context loader. Re-run the benchmark after adding skills, then after adding hooks, recording each delta. Adopt each skill/hook only if it beats the prior stage or prevents a real recurring failure.

Expect: 2-3 skills and 1-3 hooks, each with a benchmark delta, adopted only if they improved the score or prevented a known failure.

Audit before moving on: Trigger each hook to confirm it fires; confirm each skill runs standalone. Fail if a hook never fires or a skill/hook was adopted with no delta and no failure-prevention reason. On fail, fix or drop it.

Step 5 — Tool layer: MCP vs CLI + retrieval

Goal: Pick the leanest tool integration and add query-before-read.

Optimise the tool layer. (1) For each external capability you need (git, deploys, a database, scraping), compare MCP server vs a CLI tool and keep the LEANER one on the benchmark (chase.h.ai repeatedly shows CLI can beat MCP on tokens/latency). (2) Add Graphify: build the knowledge graph for this repo (`/graphify .`) and add a CLAUDE.md rule to query the graph before opening raw files; for a large/markdown repo consider LightRAG/RAG-Anything. Re-run the benchmark, especially on tasks that involve reading across many files, and record the token delta from query-before-read. Keep MCP only where it beat the CLI.

Expect: A per-capability MCP-vs-CLI decision backed by benchmark, Graphify (or LightRAG) wired with a query-before-read rule, and a token delta.

Audit before moving on: Confirm the kept integration is actually the leaner one per the benchmark, and that the agent queries the graph before raw reads. Fail if MCP was kept on preference not numbers, or the graph is built but never queried. On fail, switch to the leaner tool or enforce the query rule.

Step 6 — Parallel worktrees

Goal: Measure multi-session throughput on a real batch.

Adopt parallel worktrees the way Cherny's team does, and measure it. Take a real batch of 3-5 independent tasks on this project. Run them (a) serially in one session and (b) across 3-5 git worktrees each with its own Claude session, named and aliased for fast switching. Measure total wall-clock and any merge-conflict cost for both. Report whether parallel worktrees actually raised throughput for this batch or just added coordination overhead. Only recommend the worktree workflow if the measured throughput gain is real.

Expect: A serial-vs-parallel throughput measurement on a real batch, with merge-cost noted and a kept/not-kept recommendation.

Audit before moving on: Confirm the tasks were genuinely independent (no shared files) and both arms are real timings. Fail if worktrees were recommended without a measured gain, or tasks shared files (invalid parallelism). On fail, re-pick independent tasks or report that parallelism did not help here.

Step 7 — Subtraction pass + cost controls

Goal: Ablate stale rules and apply token controls, keeping only what helps.

Run the subtraction pass - mastery is deletion as much as addition. (1) Ablate: remove each CLAUDE.md rule group and each skill one at a time, re-run the benchmark, and DELETE anything whose removal does not hurt the score and is not a safety/money guardrail (cooper.simson's purge; Cherny's simple mode showing over-instruction caps capability). (2) Cost controls: test concise/caveman output rules and advisor mode (Opus plans, Sonnet/Haiku execute) on the benchmark; keep them if they cut tokens/cost without hurting correctness (the concise-output study chase.h.ai cites). Report the setup before and after the purge: line count, token cost, correctness. The goal is the smallest setup that holds the score.

Expect: An ablation table (rule/skill, score with/without, kept/cut), a trimmed setup, and a cost-control delta from concise output + advisor mode.

Audit before moving on: Confirm each cut was decided by a real re-run, and that concise mode did not drop correctness. Fail if nothing was cut (no real ablation) or if a token saving came with a correctness regression. On fail, actually run the ablations.

Step 8 — Validate + inheritable setup + gap report

Goal: Prove the setup beats baseline and reproduces on a second project.

Final validation. (1) Run the full benchmark on the final, trimmed setup vs the step-1 baseline and report the deltas (tokens, wall-clock, correctness). (2) Inheritance: copy the setup (CLAUDE.md template, skills, hooks, settings) to a second project (or a clean clone) and confirm it loads and behaves as intended with minimal edits. (3) Enumerate claims, mark verified/unverified/false, list missing pieces ranked by impact, fix the top two. Commit the setup; print the commit hash. Produce the full progression table: stage -> tokens/wall-clock/correctness, so the improvement (or regression) at each step is visible.

Expect: A full progression table, a second-project inheritance result, a claim ledger, and a committed setup with hash.

Audit before moving on: Confirm the final setup beats baseline on at least one metric and that the second project actually loaded it. Fail if the final setup is not better than bare, or inheritance was not tested. On fail, report honestly - if mastery amounted mostly to subtraction, say so.

4 · Gap analysis and audit

Run this after the build. It makes the agent enumerate its own claims, mark what is verified, rank the gaps and fix the top ones.

Run after the setup is built. Enumerate every claim as a numbered list: the baseline, each stage's benchmark delta, the MCP-vs-CLI decisions, the Graphify query-before-read saving, the worktree throughput, the ablation cuts, and the cost-control deltas. Mark each VERIFIED (with real run logs), UNVERIFIED, or FALSE. Specifically re-check: are all numbers measured not vibed (the session-log study's warning); was any component adopted without beating the prior stage; are stale rules written for older models still present; did concise/caveman output cut tokens without hurting correctness; is any kept MCP server actually leaner than its CLI; did parallel worktrees show a real throughput gain on independent tasks; did the second project actually inherit the setup. List missing pieces: unablated rules, unmeasured stages, MCP kept on preference, untested inheritance. Rank by impact on real productivity and cost. Fix the top three and re-verify. Report what you could not fix and why. Do not mark any stage an improvement without a measured before/after.

Questions the agent must ask itself before it says done:

5 · What proves it right

6 · Failure modes

7 · Agent direction (unattended runs)

Reach first for a bare-config benchmark on 3-5 real tasks and the project's transcripts; the whole pack hinges on measured before/after, so never start adding components before the baseline exists. Cache the baseline table and re-run it after each stage - one component at a time, or you cannot attribute the delta. Use Graphify to query before reading raw files, and prefer a CLI over an MCP server wherever the benchmark says it is leaner. Run the subtraction pass for real: ablate each rule/skill and cut what does not earn its place, because stale rules drag newer models down. Spawn sub-agents for the parallel-worktree throughput test (that is the point) but only on genuinely independent tasks. Stop and ask the user only if there is no git history or no real tasks to benchmark. Return a markdown progression report (stage-by-stage table, ablation results, cost delta, worktree result, 'Unverified and missing') plus the committed setup: CLAUDE.md, primer.md, tasks/lessons.md, skills, hooks, settings. Print the commit hash. If mastery turned out to be mostly subtraction, report that honestly.

Prompt system updated 2026-10-22 · all prompt systems · JSON

Browse more deep dives at DejaViewed — 846 entries across the full catalog.

Generated 2026-10-22 · curated by @6ab3 for DejaViewed