Page 91 of 97 — MCP servers, Claude Code skills, prompt packs, and workflows.
Feed Claude a CSV of price data — returns quant-style analysis: trend strength, volatility regime, MA crossovers, and explicit entry/exit rules. Turns Claude into a CSV-first trading analyst.
Claude builds a risk-management system for daily-profit traders — position sizing, drawdown limits, and blowup avoidance rules. The gate that sits between idea and execution.
Claude maps how smaller daily gains scale into larger monthly income via disciplined compounding — produces a realistic growth curve and the behavioral rules that protect it.
Akshay Pachaar's deep tutorial on configuring the .claude/ folder end-to-end: CLAUDE.md vs CLAUDE.local.md (project vs personal), the rules/ folder for path-scoped modular instructions, hooks/ for deterministic control (PreToolUse/PostToolUse/Stop with exit-code 2 as the block gate), skills/ for auto-invoked workflows with YAML frontmatter, agents/ for specialized subagent personas with restricted toolsets, and settings.json for allow/deny permission policies. Covers both project-level and global ~/.claude/ setups.
One GitHub archive that scraped every Claude/Claude-Code repo worth knowing about — 1,900+ projects indexed so you can skip the "reply to get access" DM funnels and just browse. Cole Manion's bot did the collection work so yours doesn't have to.
Seven-step playbook for a solo founder to go from zero to $30K/month with Claude driving the build: Open SaaS for the foundation, Supabase for data, Repomix to give Claude full project context, pre-built Anthropic Skills for integrations, Flowise/LangChain for AI features, Vercel for deploy, Stripe for payments. Comes with the math: 100 users × $29/mo = $2.9K, 350 users = $10K — the hard part isn't building anymore, it's picking a real problem.
Akshay Pachaar's one-sheet reference covering 50+ Claude Code commands, grouped by what you're actually trying to do: project setup & memory, context & session lifecycle, code review & quality, CLI flags, monitoring & reporting, system config & diagnostics, model & thinking controls, inline prefixes, keyboard shortcuts, input/interaction modes, and multi-agent automation.
ComposioHQ's 9,000-plus indexed Claude plugin list — comes with adoption metrics so you can see what people actually use in production vs. what has 3 commits and 2 stars.
Autonomous coding-agent framework powered by Claude — Anton Osika's take on an end-to-end agent that plans, writes, runs, and revises code without a human driving each turn.
Runs Claude Code itself as an MCP server — enables the agent-in-agent pattern. Claude Code can recursively delegate coding tasks to nested Claude Code instances. Integrates with Claude Desktop and other MCP clients.
23,000+ lines of docs, 219 templates, 271 quizzes — the beginner-to-power-user curriculum for Claude Code. FlorianBruniau's companion repo to the community-bible awesome-list.
Definitive prompt-engineering guide for Claude Code mastery covering Opus 4.6, Sonnet 4.6, Haiku 4.5 — includes Skills, Plugins, MCP, Hooks, Agent Teams, Ultrathink. 100+ curated skills and 220+ verified sources.
Curated library of high-quality Claude prompts covering writing, coding, analysis, research, and more. Maintained by langgptai — pull a battle-tested prompt instead of reinventing one every session.
50+ interface styles, 161 color palettes, and 99 UX guidelines Claude can reference when generating UI. Claude finally stops shipping the same 3 ugly layouts on every project.
Nikiforovall's handbook of best practices and techniques for Claude Code workflows — focused on daily-use patterns rather than theoretical best-practice lists. Ships distributable plugins alongside the prose.
Comprehensive project-management workflow for Claude Code, built by Ran Aroussi. Many specialized agents plus slash commands with strong documentation — turnkey PM stack instead of cobbling your own from scratch.
2.2K-star collection of 100+ production-ready subagents covering comprehensive domain expertise. davepoon's drop-in library — plug into any Claude Code workflow without rolling your own.
Advanced Claude Code workflows and real-world use cases collected by joe-njenga — different curator angle than the affaan-m hackathon edition, leaning more toward daily-use patterns.
Curated agent skill sets usable across Claude and other LLMs — VoltAgent's pick list that doesn't assume any single model vendor. Useful if your stack routes between Claude, GPT, and open-source.
Graph-plus-vector RAG system that helps Claude understand large codebases structurally rather than as a soup of chunks. Gives Claude topology awareness — which files depend on which, where the real entry points are.
Massive 25K-star collection of production-ready subagents covering strategy, dev, security, design, data, and research. Drop-in library for Claude Code workflows — don't hand-roll what wshobson already wrote.
3.7K-star team-of-specialized-agents pattern — agents collaborate on full features, each carrying expert-level knowledge across a specific stack. vijaythecoder's take on multi-agent Claude orchestration.
MCP server that lets you talk to Claude with real voice — Whisper for speech-to-text in, Kokoro for text-to-speech out. Full voice round-trip without leaving Claude Code.
5.6K-star terminal multiplexer for AI coding agents — manages Claude Code, Aider, Codex, OpenCode, and Amp in one UI so you can run parallel agents without the window-switching chaos.
Jesse Vincent's (obra) Superpowers repo — 160K stars, 47 skills that give Claude capabilities it doesn't ship with: web browsing inside sessions, persistent memory, image generation, auto-summarization, file watching with live reload, shell exec with safety checks. Install = `cp -r superpowers/skills ~/.claude/skills/`.
Long-term memory for Claude Code via compression — persists context between sessions, stays lightweight, and fixes the #1 thing everyone complains about: "why does Claude forget my codebase every session?"
20.9K-star tool that packs your entire codebase into one AI-readable file — XML, Markdown, or plain. Perfect for feeding big repos into Claude's context window or any LLM that consumes a single-file bundle.
Curated plugins, extensions, and memory systems for Claude — skews toward persistent-memory patterns and long-running agent setups that need state to survive between sessions.
75+ workspace templates for building Claude agents — goes beyond coding into research, ops, legal, content, finance. Each blueprint is a full directory structure, skill set, and system prompt starter.
FlorianBruniau's variant of awesome-claude-code — the community bible leaning into curated skills, hooks, slash commands, and orchestrators. Different curator than hesreallyhim's index; both worth bookmarking.
MCP server that bridges Claude Code to n8n's 400+ integration catalog — Claude gains read/write access to Gmail, Notion, Slack, Salesforce, Stripe, etc., through n8n's existing connector library.
Turns any codebase or knowledge base into an interactive knowledge graph you can explore, search, and query. Works with Claude Code, Codex, Cursor, Copilot, and Gemini CLI — insane for onboarding into big repos.
Broad-ecosystem awesome-list for Claude — tools, apps, and resources that go beyond Claude Code proper. Sits next to the developer-centric awesome-claude-code index as the wider "everything Claude touches" directory.
Orchestration framework for running many Claude Code agents in parallel — systematically improves your codebase at scale with automation-first design. Dicklesworthstone's answer to "how do I run 20 Claudes at once without chaos?"
Hackathon-winning Claude Code harness (Cerebral Valley x Anthropic) built from 10+ months of daily use: 30 agents, 136 skills, 60 slash commands, 14 MCP configs, 1,282 tests with 98% coverage, 102 static-analysis rules. Turnkey production-quality setup.
Official plugin marketplace for the Superpowers ecosystem — 20+ battle-tested skills including /brainstorm, /write-plan, /execute-plan commands and sessionStart context injection. Companion storefront to the obra/superpowers base repo.
Community-curated collection of reusable Claude skills and automations — pick a skill folder, drop it into ~/.claude/skills/, and Claude inherits the capability.
1,200+ ready-to-use Claude skills — one of the largest single-repo skill collections that exists. Bulk-install or cherry-pick from the catalog.
2.4K-star background agent that whispers guidance to Claude Code — watches sessions, reads files, builds memory over time, and surfaces relevant context when you need it. 0xfurai's "second brain for Claude."
MCP server by zilliztech for semantic code search across massive codebases — pulls the right files without manual breadcrumbing. Built for huge repos where default context fails and glob/grep isn't enough.
Command-line tool that tracks exactly how much every Claude Code session costs you — per session, per project, per model. Replaces the "how much did I burn last week" guesswork with real numbers you can actually budget around.
Always-current mirror of the Claude Code docs, structured so Claude itself can read them at runtime. Point Claude at the repo and it can look up its own CLI, hooks, skills, MCP, and routines docs without you copy-pasting from the web.
The 20 paid Claude features most subscribers never use, ranked by how much they change your workflow: Projects, custom instructions per project, project files, extended thinking, code execution, file uploads, vision, web search, artifacts, data analysis mode, file creation, voice mode, GitHub integration, Drive integration, Zapier/tool connectors, Claude Code (included in Max), multi-model access, priority new-feature access, new-chat-per-topic discipline, and shareable artifact links.
Opinionated Claude Code framework that bundles 30+ slash commands and a library of personas so Claude reasons and responds like a senior dev on your team. Install once, and /architect, /security-review, /refactor, /profile all become one-liners.
Turns freeform product specs into structured tasks Claude Code actually follows. Instead of giving Claude a vague "build feature X" prompt, Task Master decomposes the spec into ordered, testable units that Claude can work through sequentially without drifting.
Ready-to-install pack shipping 600+ agents and 200+ slash commands for Claude Code. One install command pulls in battle-tested agent definitions covering coding, research, ops, and content — no hand-rolling your own subagent configs.
19.7K-star, #1-trending GitHub Claude Code playbook from Anthropic insiders. 84 battle-tested tactics: Command→Agent→Skill orchestration, parallel agent teams with tmux + git worktrees, the Ralph Wiggum autonomous-loop pattern, cross-model workflows (Claude reviewing Codex output), Opus-plans → Sonnet-executes splits, agentic search over RAG, spawn-subagents-don't-compress-context.
The community-maintained master index of MCP servers for Claude — Slack, Gmail, GitHub, databases, knowledge bases, and everything in between. One-stop shop when you need to connect Claude to a new tool.