Claude

87 sources · Updated March 27, 2026
Claude Code is now a $2.5B run-rate product powering 4% of all GitHub commits, and the centerpiece of a paradigm where agents replace manual coding as the primary implementation layer. The workflow has matured beyond Research-Plan-Build into a general work operating system: CEOs use it as an AI Chief of Staff (unifying inboxes, overnight todo lists), founders use it for personalized capacity planning on Linear, and outbound sales teams run 11-API pipelines through Skills files. Plan files and /handover commands preserve context across sessions; Taskmaster forces long-running sessions to complete; specialized harnesses (designer, marketer, sales) compound value as skills accumulate.

Sub-Topics

14 sources

Claude: Architecture & Knowledge Patterns

The dominant second brain architecture in 2026 is vault as foundation, Claude Code as engine — a structured local knowledge base of plain text markdown that any AI agent can read directly. The repo-as-database pattern means quality of knowledge organization directly impacts quality of AI reasoning, and "Personal Context Management" (Forte Labs) reframes the bottleneck: it's not AI capability, it's context delivery. Exo Brain advances this with a key insight: maintenance kills PKM systems, so make Claude do the maintenance via self-feeding read/write loops, while the "agents read, humans write" principle keeps vault content purely human-authored for honest pattern detection. The file-system-as-database approach is agent-agnostic and extends to self-updating knowledge bases via scheduled crawl-to-markdown pipelines. Claude Code was born from a belief in terminal simplicity as the right AI interface, drawing parallels to TypeScript's adoption curve. Cost architecture matters: 80% of agent tasks are janitorial and don't need frontier intelligence, making hierarchical model routing (10x cost reduction) essential at scale. The scratchpad/napkin pattern provides a distinct form of agent memory that compounds across sessions — agents that log their own mistakes exhibit improving performance over time. Obsidian + Claude Code skills create a persistent knowledge system where /skills map naturally to the file-based architecture.

5 sources

Claude: Design

Design with Claude Code operates through a 3-layer harness: Skills for expertise (Impeccable, Emil Kowalski, Interface Design), Agent Canvases for design surfaces (Paper with code-native HTML/CSS, Pencil with 6-agent swarm mode), and Inspiration tools for taste (Variant Style Dropper, Mobbin, Cosmos). Multimodal input is a game-changer — recording video of a target UI and feeding it through Claude produces better results than text prompts. The design-to-code gap is closing from both sides: Figma-like visual editors let users select front-end elements visually and apply edits through Claude Code, while the Visual Explainer skill transforms dense terminal output into rich HTML pages with consistent design via CSS pattern libraries. Skills that control output format (not just task execution) represent a new category of agent customization. The component library ecosystem is actively building for agents: shadcn/cli v4 introduced "shadcn/skills" as a first-class concept.

34 sources

Claude: Ecosystem & Tools

The Claude Code ecosystem is maturing at an extraordinary pace, now backed by $2.5B in Claude Code revenue (run rate) and 4% of all GitHub commits. Anthropic itself is investing heavily: 11 open-sourced enterprise plugins, a free official course, and unreleased features including Agent Teams and Thinking Edits in the pipeline. Skill distribution has formalized: gstack established skill packs as "dotfiles for AI," skills install via npx, Josh Pigford shipped a dedicated macOS app, and self-learning skills that create new skills from usage patterns represent the frontier. Enterprise adoption is real: Intercom built 13 plugins with 100+ skills, $1M+ customers 40x'd year-over-year to 500+, and agent-first teams generate 30+ internal tools in weeks. Skill quality is becoming systematic: built-in test generation for trigger rate, autoresearch for continuous eval loops, and state-machine modeling for complex skills. The component library ecosystem is building for agents (shadcn/cli v4 with first-class skills). The Obsidian + Claude Code stack is the recognized community pattern for second-brain workflows. Developer tooling is flourishing: CodexBar for token cost tracking, Decode for browser+whiteboard integration, X search skills for real-time intelligence, and community-curated resource repos. Claude Code's reach extends beyond engineering via /loop patterns and 1,200+ hour research workflows.

18 sources

Claude: Settings & Config

Claude Code configuration operates at three layers: environment (bypass permissions via --dangerously-skip-permissions, Zed autosave at 500ms, Ghostty + Zed split-screen, game sound hooks for task notifications), context architecture (CLAUDE.md under 200 lines with one-liner declarative behaviors, .claude/rules/ with path-scoped YAML, _MANIFEST.md with Tier 1/2/3 priority loading), and skill design (progressive disclosure cutting 89% context bloat, state-machine skills with data dependency optimization). The platform has six distinct extension mechanisms: Plugins (packaged workflows), Skills (on-demand knowledge), MCPs (external tool connections), Commands (prompt shortcuts), Subagents (delegated tasks), and Hooks (pre/post action scripts). Web fetch uses a subagent architecture that summarizes pages to prevent context overflow but can lose detail. The biggest insight is that system engineering beats prompt engineering — investing 2 hours in context architecture lets subsequent prompts be 10 words long. An 8-phase bootstrap prompt can build an entire workspace from scratch. Effort levels (low/medium/high) are configurable and sticky; conflicting CLAUDE.md instructions are the most common cause of degraded output at high effort.

2 sources

Claude: Voice & Mobile

Voice input tools, dictation workflows, remote access patterns, and voice cloning. Monologue or WhisperFlow pipe speech into apps while Claude fills transcription gaps, Mac Mini + Telegram enables mobile Claude Code access, and Sonnet 4.5 can clone individual writing voices with 94% blind-test accuracy across literary, essay, and professional styles -- enabling executives to delegate communication while maintaining authentic voice.

38 sources

Claude: Workflow

Claude Code workflows have evolved far beyond the original Research-Plan-Build loop into a general work operating system. The core pattern remains plan-first — plan.md files survive context loss, /handover commands preserve institutional knowledge across sessions, and 4-6 parallel Ghostty sessions enable concurrent work. But the frontier has moved to agent-first engineering: teams deleting their IDEs, CEOs using Claude as an AI Chief of Staff that unifies 6+ inboxes and manages overnight todo lists, and multi-month projects shipping in a day. Specialized harnesses (designer, marketer, sales) compound as skills accumulate, with the outbound sales use case (11 APIs, 72 scripts) demonstrating Claude Code as a full business automation platform. Reliability infrastructure is emerging: Taskmaster forces long-running sessions to complete, audio notifications signal task completion for background delegation, and the "Claude Code to Agent SDK" pipeline moves prototyped workflows into headless production. Beyond engineering, Claude Code serves as an intelligent query layer on top of Linear for personalized capacity planning, a 5-API meeting prep pipeline, a content strategy tool, and even PowerPoint generation. Cowork with GSuite/Slack connectors and /loop for recurring business tasks extend the reach to non-technical professionals.