llms.txt
+66 −26
https://skills-hub.ai/llms.txt
− > The open registry for AI coding skills — install, version, and compose 4,400+ SKILL.md files in one command. Works with Claude Code, Cursor, Codex CLI, Windsurf, GitHub Copilot, and any MCP-compatible tool.− - [Browse 4,400+ skills](https://skills-hub.ai/browse): Filter by category, platform, source, tag, SDLC phase, quality score, or security grade.− - [Kimi K2.7 Code in GitHub Copilot: The First Open-Weight Model in a Mainstream AI Coding Assistant](https://skills-hub.ai/blog/kimi-k2-7-code-github-copilot-2026): Moonshot AI's Kimi K2.7 Code landed in GitHub Copilot July 1 2026 — first open-weight, MIT-licensed model in the five-lab model picker. 81.1% MCPMark, 32B active params (1T total MoE), GPT-5.4 mini tier pricing, hosted on Azure. Task routing guide, admin setup for Business/Enterprise, open-weight compliance checklist.− - [GPT-5.6 Sol, Terra, and Luna: The Developer's Tier-Selection Guide](https://skills-hub.ai/blog/gpt-5-6-sol-terra-luna-coding-guide-2026): OpenAI's GPT-5.6 (June 26, 2026) ships as Sol ($5/$30), Terra ($2.50/$15), and Luna ($1/$6) per 1M tokens — with Sol ultra mode for agentic subagent reasoning and explicit cache breakpoints for 80%+ input cost savings. Tier-selection decision tree, pricing math, and Codex CLI wiring guide.− - [Claude Code's July Pivot: Background Agents Deliver PRs, Manual Mode Is the New Default](https://skills-hub.ai/blog/claude-code-auto-pr-manual-mode-july-2026): Six releases in eight days (v2.1.197–204, July 1–8 2026) — background agents now auto-commit, push, and open draft PRs when worktree tasks finish; Manual is the enforced default permission mode; dynamic workflow sizing added; 25+ background-agent reliability fixes.− - [X Ships an Official MCP Server: What XMCP Means for Your AI Coding Workflow](https://skills-hub.ai/blog/x-xmcp-official-mcp-server-2026): X launched XMCP on June 30 2026 — a hosted MCP server at api.x.com/mcp with 200+ tools auto-generated from the X API OpenAPI spec. Setup guide for Claude Code and Cursor, what you can query, pricing, and four production-ready agentic workflows.− - [claude mcp login & /rewind: The Two Claude Code June 2026 Features Worth Your Attention](https://skills-hub.ai/blog/claude-code-mcp-login-rewind-2026): Claude Code v2.1.185–v2.1.193 (June 22–26) shipped OAuth-native MCP server auth with `claude mcp login` and session context recovery with `/rewind`. Setup guide, token lifecycle, and daily workflow patterns.− - [MCP 2026-07-28: Six Breaking Changes and the New MCP Apps Primitive](https://skills-hub.ai/blog/mcp-spec-july-28-apps-migration-2026): The largest MCP revision since launch ships July 28. Sessions are gone, six things will break in production servers, and MCP Apps (SEP-1865) lets tools ship interactive HTML UIs to any host. Migration checklist and MCP Apps primer inside.− - [Claude Sonnet 5: The Agentic Upgrade That Changes Your Daily Workflow](https://skills-hub.ai/blog/claude-sonnet-5-agentic-coding-2026): Claude Sonnet 5 (June 30 2026) — 80.5% on Terminal-bench 2.1 (+13.5pp over Sonnet 4.6), configurable reasoning effort levels, $2/M input introductory pricing. What changes for Claude Code users, how to configure effort levels, and when to keep Opus 4.8.− - [Claude Enterprise-Managed Auth: Zero-Touch MCP Connectors via Okta](https://skills-hub.ai/blog/claude-mcp-enterprise-auth-okta-2026): Anthropic's June 18 2026 feature — admin provisions Okta once, employees inherit Asana, Atlassian, Figma, Linear, and four more MCP connectors automatically across Claude Chat, Claude Code, and Cowork. No per-user OAuth.− - [Gemini 2.5 Pro Deep Think: What 2M Tokens Means for AI Coding](https://skills-hub.ai/blog/gemini-2-5-pro-deep-think-coding-2026): Google's June 22 2026 release — 87.6% on LiveCodeBench, 2M token context window, visible Deep Think reasoning chains. What actually changes for developers building with AI: whole-codebase analysis, security audits across hundreds of files, and how it stacks up against Claude Fable 5.− - [Cascade is Dead: Migrate to Devin Local Today Before Your CI Breaks](https://skills-hub.ai/blog/cascade-eol-devin-local-migration-2026): Cascade (the original Windsurf local agent) is EOL July 1 2026. In-editor migration is automatic; CI scripts calling Cascade by name will break silently. Migration playbook, grep commands, ACP integration guide, and what Devin Local (Rust rewrite, 30% more token-efficient, subagent support) adds over Cascade.− - [ACP Spec 1.0: Build Your AI Agent Once, Run It in Every IDE](https://skills-hub.ai/blog/acp-spec-1-open-standard-2026): ACP Spec 1.0 is the first stable, open-governance release of the Agent Client Protocol — the LSP for AI coding agents. 25+ compatible agents including Claude Code, Codex, Devin Local, and OpenCode. JSON-RPC 2.0 wire protocol, capability negotiation, and Spaces context sharing. Setup guide with agents.json examples.− - [Gemini CLI Enterprise Pivot 2026](https://skills-hub.ai/blog/gemini-cli-enterprise-pivot-2026): Google closed Gemini CLI's free tier on June 18 2026 after 6,000 community PRs — migration guide and alternatives.− - [Kiro 0.12: Parallel Tasks and Quick Plan Mode](https://skills-hub.ai/blog/kiro-parallel-tasks-quick-plan-2026): Kiro 0.12 (May 2026) ships parallel task execution (4x faster), Quick Plan Mode (spec-to-code in a single pass), and pre-implementation requirements analysis.− - [Kiro Requirements Analysis: SMT Solver for AI Slop-Free Specs](https://skills-hub.ai/blog/kiro-requirements-analysis-smt-2026): Deep dive on Kiro's neurosymbolic Requirements Analysis engine — three-stage pipeline (LLM rewrite → formal logic → SMT proof) that catches logical contradictions in specs before code generation begins.− - [OpenAI Codex and GPT-5.5 on Amazon Bedrock: The Developer Playbook](https://skills-hub.ai/blog/openai-codex-aws-bedrock-2026): GPT-5.5, GPT-5.4, and Codex went GA on Amazon Bedrock June 1 2026 — auth wiring, VPC isolation, Responses API migration, and per-token cost modeling.− - [Claude Code Dynamic Workflows 2026](https://skills-hub.ai/blog/claude-code-dynamic-workflows-2026): v2.1.152–154 (May 27–29) ships dynamic workflows for 100+ background agents, MessageDisplay hooks, and hot-reloadable skills.− - [GitHub Copilot Browser Tools Are GA: AI Agents Now Verify Their Own Front-End Work](https://skills-hub.ai/blog/github-copilot-browser-tools-ga-2026): Browser tools hit GA in Copilot for VS Code on July 1 2026 — agents navigate, interact, and screenshot the running app mid-task. What changed from research preview, how to enable it, and how to replicate the workflow in Claude Code and Cursor via the new browser-agent-verify SKILL.md.− - [Best Claude skills (2026)](https://skills-hub.ai/blog/best-claude-skills-2026): Top 25 most-installed Claude skills, ranked.− - [GitHub Copilot Flex Billing 2026](https://skills-hub.ai/blog/github-copilot-flex-billing-2026): Complete playbook for the June 1 AI Credits transition — plans, credit math, budget controls, and 10-day checklist.− - [Windsurf 2.0: Agent Command Center, Spaces & Devin Cloud](https://skills-hub.ai/blog/windsurf-2-agent-command-center-2026): Deep dive on Windsurf 2.0 (April 15, 2026) — Agent Command Center Kanban, Spaces for persistent task context, and one-click Devin cloud handoff.− - [Cursor Composer 2.5: Build in Parallel, 10x Cheaper](https://skills-hub.ai/blog/cursor-composer-2-5-2026): Deep dive on Cursor Composer 2.5 (May 18, 2026) — Agent Swarm parallel architecture, targeted RL training, 79.8% SWE-bench at one-tenth the cost of Opus 4.7.− - [Google Antigravity 2.0 SDK: Build Custom Agents on Your Own Infrastructure](https://skills-hub.ai/blog/google-antigravity-2-sdk-2026): Everything developers need to know about the Antigravity 2.0 SDK, agy CLI, Managed Agents API, and multi-agent orchestration launched at Google I/O 2026.− - [Amazon Kiro: The Spec-Driven IDE That Takes Prototypes to Production](https://skills-hub.ai/blog/amazon-kiro-spec-driven-ide-2026): How AWS Kiro's spec-first workflow, EARS notation, agent hooks, and AWS MCP Server GA change the AI IDE landscape in 2026.− - [Sources](https://skills-hub.ai/sources): 90+ official sources — Anthropic, Microsoft, Google, Vercel, Stripe, Cloudflare, and more.+ > The open registry for AI coding skills — install, version, and compose 13,000+ SKILL.md files in one command. Works with Claude Code, Cursor, Codex CLI, Windsurf, GitHub Copilot, and any MCP-compatible tool.+ - [Browse 13,000+ skills](https://skills-hub.ai/browse): Filter by category, platform, source, tag, SDLC phase, quality score, or security grade.+ - [Sources](https://skills-hub.ai/sources): 270 official sources — Anthropic, Microsoft, Google, Vercel, Stripe, Cloudflare, and more.+ ## Blog posts+ Newest first. Machine-readable: [JSON Feed](https://skills-hub.ai/blog/feed.json), [Atom](https://skills-hub.ai/blog/feed.xml).+ - [OpenAI Codex CLI Updates: Instant Interrupts, Guardian Reviews, and Plugin Caching](https://skills-hub.ai/blog/openai-codex-cli-recent-releases-2026): The latest OpenAI Codex CLI releases introduce instant interrupt steering, opt-in Guardian reviews for agent handoffs, and performance improvements through plugin manifest caching. Here is what matters for developers writing and using AI coding skills.+ - [Name a Skill verify and Claude Code Will Run It Before Committing](https://skills-hub.ai/blog/claude-code-verify-skill-name-convention-2026): Claude Code v2.1.286 (September 30, 2026) made one skill name load-bearing: if your project or user skills include one called verify, Claude is now told to run it right before committing. Two other folder names silently refuse to load, and the new /doctor prompt-audit reads your skills for instructions written for older models.+ - [GitHub Copilot Browser Tools Are GA: AI Agents Now Verify Their Own Front-End Work](https://skills-hub.ai/blog/github-copilot-browser-tools-ga-2026): As of July 1, 2026, GitHub Copilot's browser tools hit general availability in VS Code — letting AI coding agents drive a real browser to verify their own front-end changes. Here's what changed, how it works, and how to apply the same pattern in Claude Code and Cursor.+ - [Kimi K2.7 Code in GitHub Copilot: The First Open-Weight Model in a Mainstream AI Coding Assistant](https://skills-hub.ai/blog/kimi-k2-7-code-github-copilot-2026): Moonshot AI's Kimi K2.7 Code landed in GitHub Copilot on July 1 — the first open-weight model in the five-lab model picker. MIT-licensed weights on Hugging Face, hosted on Azure, billed at GPT-5.4 mini tier pricing. Here's what it means for your workflow.+ - [Claude Code's July Pivot: Background Agents Deliver PRs, Manual Mode Is the New Default](https://skills-hub.ai/blog/claude-code-auto-pr-manual-mode-july-2026): Six releases in eight days (v2.1.197–204) shipped Claude Code's biggest behavioral shift of 2026: background agents now commit, push, and open draft PRs autonomously when worktree tasks finish; 'default' permission mode was renamed Manual and enforced across every surface; and dynamic workflow sizing lets you tune agent-pool scale from+ - [GPT-5.6 Sol, Terra, and Luna: The Developer's Tier-Selection Guide](https://skills-hub.ai/blog/gpt-5-6-sol-terra-luna-coding-guide-2026): OpenAI's GPT-5.6 ships as three distinct models — Sol (frontier agentic), Terra (balanced daily driver), and Luna (fastest/cheapest). Here's exactly when to use each tier in your AI coding stack, with pricing math and real workflow patterns.+ - [X Ships an Official MCP Server: What XMCP Means for Your AI Coding Workflow](https://skills-hub.ai/blog/x-xmcp-official-mcp-server-2026): On June 30, X launched XMCP — a hosted MCP server at api.x.com/mcp with 200+ tools auto-generated from the X API OpenAPI spec. Here is how to wire it into Claude Code, Cursor, or any MCP-compatible agent, what you can query, and why every major platform will follow.+ - [claude mcp login & /rewind: The Two Claude Code June 2026 Features Worth Your Attention](https://skills-hub.ai/blog/claude-code-mcp-login-rewind-2026): Claude Code v2.1.185–v2.1.193 (June 22–26, 2026) shipped two underrated features: OAuth-native MCP server authentication with `claude mcp login`, and /rewind for recovering context after /clear. Here's what they do and how to use them today.+ - [MCP 2026-07-28: Six Breaking Changes and the New MCP Apps Primitive](https://skills-hub.ai/blog/mcp-spec-july-28-apps-migration-2026): The largest MCP revision since launch ships July 28. Sessions are gone, six things will break in your server, and a brand-new primitive — MCP Apps — lets tools ship interactive HTML UIs to any host. Migration checklist and MCP Apps primer inside.+ - [Claude Sonnet 5: The Agentic Upgrade That Changes Your Daily Workflow](https://skills-hub.ai/blog/claude-sonnet-5-agentic-coding-2026): Claude Sonnet 5 shipped June 30 with a 13-point Terminal-bench jump, configurable reasoning effort, and introductory pricing that undercuts Opus for most automation. Here's what changes for Claude Code users today.+ - [Claude Enterprise-Managed Auth: Zero-Touch MCP Connectors via Okta](https://skills-hub.ai/blog/claude-mcp-enterprise-auth-okta-2026): Anthropic shipped enterprise-managed authorization for MCP connectors on June 18, 2026. Admins provision Okta once; employees get Asana, Atlassian, Figma, Linear, and four more connectors on first login across Claude Chat, Claude Code, and Cowork — no per-user OAuth.+ - [Cascade is Dead: Migrate to Devin Local Today Before Your CI Breaks](https://skills-hub.ai/blog/cascade-eol-devin-local-migration-2026): Cascade reaches end-of-life on July 1, 2026. In-editor workflows migrate automatically — but CI scripts and automations that call Cascade by name will break silently. Here's the exact grep, the exact fix, and what Devin Local adds over what you're replacing.+ - [Gemini 2.5 Pro Deep Think: What the 2M-Token Context Means for AI Coding](https://skills-hub.ai/blog/gemini-2-5-pro-deep-think-coding-2026): Google released Gemini 2.5 Pro with Deep Think on June 22, 2026 — 87.6% on LiveCodeBench, 2M token context, and visible reasoning chains. Here's what actually changes for developers who build with AI.+ - [ACP Spec 1.0: Build Your AI Agent Once, Run It in Every IDE](https://skills-hub.ai/blog/acp-spec-1-open-standard-2026): The Agent Client Protocol just hit its first formal spec release under open governance. ACP is JSON-RPC 2.0 over stdin/stdout — the LSP for AI agents — backed by Zed, JetBrains, GitHub, Google, and Cognition. Here is what changed, which tools are compatible, and how to wire Claude Code, Codex, and Devin Local into a single unified command center.+ - [Fable 5 Was Pulled in 72 Hours: What the Export-Control Suspension Means for Your AI Stack](https://skills-hub.ai/blog/fable-5-export-control-suspension-2026): On June 12, US export-control regulations forced Anthropic to suspend Claude Fable 5 globally — just 72 hours after launch. Here is the migration playbook, the four codebase dependencies you need to audit, and the model-agnostic architecture that makes the next suspension a non-event.+ - [Arcade's $60M Bet on Agent Authorization: Why Production AI Agents Need a Permission Layer](https://skills-hub.ai/blog/arcade-agent-authorization-production-2026): Enterprise AI agents keep stalling after pilot. Arcade just raised $60M to fix the real bottleneck: not the model, but authorization. What the $72M company builds, why a former Okta leader is behind it, and what every developer needs in a production agent stack.+ - [Claude Code Artifacts: Turn Any Coding Session Into a Live, Shareable Page](https://skills-hub.ai/blog/claude-code-artifacts-live-pages-2026): On June 18, Anthropic launched Artifacts in Claude Code — session work published as live, auto-updating web pages shareable with authenticated teammates. PR walkthroughs, incident timelines, system explainers, dashboards, and release checklists, all from your terminal.+ - [Kiro IDE 1.0: Agent Focus Mode, Custom Agents, and the Permission System That Audits Every Tool Call](https://skills-hub.ai/blog/kiro-ide-1-agent-focus-mode-custom-agents-2026): Kiro shipped 1.0 today with six major features. Agent Focus Mode brings a chat-first layout for parallel sessions; custom agents are Markdown files with four tool-access tags; Natural Language Hooks translate plain English into structured JSON; and the new permission system evaluates every file read, shell command, an+ - [Claude Design + /design-sync: The Two-Way Bridge Between Design and Code](https://skills-hub.ai/blog/claude-design-sync-code-handoff-2026): Anthropic's June 17 Claude Design overhaul lands design system imports, WYSIWYG canvas editing, and a bidirectional /design-sync command. Claude Code and Claude Design now share components without screenshots or rebuilds — one million users in the first week.+ - [Cursor Teams Standard vs Premium Seats: The June 2026 Pricing Breakdown](https://skills-hub.ai/blog/cursor-teams-standard-premium-pricing-2026): Cursor split Teams into Standard ($32/mo annual) and Premium ($96/mo annual) seats this month, with two separate usage pools and new admin spend controls. Here's how to right-size your team before your July billing cycle.+ - [Codex Record & Replay: Show Your Workflow Once, Get a Reusable Skill](https://skills-hub.ai/blog/codex-record-replay-skill-capture-2026): Codex Record & Replay shipped June 18, 2026 — the first AI coding tool feature that turns a demonstrated workflow into a portable, editable SKILL.md. Here's how recording works, what the generated skill looks like, how to replay it with new inputs, and how to export it to Claude Code, Cursor, or any tool on the registry.+ - [Figma Dev Mode MCP: Let Your AI Coding Agent Read the Design Directly](https://skills-hub.ai/blog/figma-dev-mode-mcp-server-2026): Figma's official Dev Mode MCP server gives Claude Code, Cursor, and Windsurf structured access to components, variables, and Code Connect mappings — replacing screenshot guesswork with machine-readable design context. Here's the complete setup and workflow guide.+ - [Cursor 3 Cloud Agents: /in-cloud, Bugbot 3×, and the PR Babysitting Pipeline](https://skills-hub.ai/blog/cursor-3-cloud-agents-2026): Cursor 3.7 (June 17, 2026) ships isolated cloud subagents via /in-cloud, reusable environment snapshots that spin up in under 10 minutes, and /babysit for overnight PR preparation. Bugbot hit 3× faster reviews the week before. Here's how to wire it all together.+ - [Cohere North Mini Code: The Open-Source 30B Agent That Fits on One H100](https://skills-hub.ai/blog/cohere-north-mini-code-open-source-2026): Cohere's North Mini Code landed June 9 as the first Apache-licensed agentic coding model that genuinely outperforms much larger proprietary alternatives — 33.4 on the AI Coding Index, 256K context, single-H100 FP8 deployment. Here's the full developer playbook.+ - [OpenCode: The Model-Agnostic Terminal Agent That Just Hit 160K Stars](https://skills-hub.ai/blog/opencode-model-agnostic-terminal-agent-2026): OpenCode connects to 75+ AI providers from a single CLI, reads code through Language Server Protocol, and costs $2–8 per day instead of $20–40/month. Here's why 7.5M developers made the switch — and how to set it up.+ - [Claude Agent SDK Billing Split: What the June 15 Change Breaks and How to Fix It](https://skills-hub.ai/blog/claude-agent-sdk-billing-split-2026): On June 15 2026, Anthropic split Claude subscription billing into two buckets: one for interactive use, one for autonomous agents. If you run claude -p, GitHub Actions automations, or any SDK-driven pipeline, your credits now drain from a separate pool — and stop when they run out.+ - [Claude Code Week 24: fallbackModel, Safe Mode, and /cd for Production Resilience](https://skills-hub.ai/blog/claude-code-fallback-models-safe-mode-2026): Claude Code's June 2026 Week 24 update shipped four features that matter most to teams running agents at scale: fallbackModel chains, --safe-mode troubleshooting, /cd workspace switching, and nested sub-agents five levels deep.+ - [Fable 5's 1M-Token Context: A Practical Guide to Whole-Codebase Analysis](https://skills-hub.ai/blog/fable-5-million-token-codebase-2026): Fable 5 ships with a 1-million-token context window — enough to load your entire codebase without summarization tricks. Here's what actually works in practice: sizing repos, caching strategy, recall tradeoffs, and a four-phase migration playbook.+ - [Kiro CLI 2.1: Headless Mode Brings AI Code Review to Every CI/CD Pipeline](https://skills-hub.ai/blog/kiro-headless-ci-skills-slash-commands-2026): Kiro CLI 2.1 ships headless mode and skills as slash commands. Here's how to wire Kiro into GitHub Actions for automated code review, dependency audits, and PR summaries — no browser, no interactivity, no waiting.+ - [Claude Code Ultracode: The xhigh + Orchestration Mode That Rewires Your Sessions](https://skills-hub.ai/blog/claude-code-ultracode-2026): Ultracode is not just a higher effort level — it pairs xhigh reasoning with automatic parallel workflow orchestration so Claude plans and runs multi-agent pipelines on your behalf. Here's the complete guide: what it does, when to use it, and how to stay inside your token budget.+ - [Claude Fable 5 in Claude Code: The Mythos-Class Upgrade for Real Coding Work](https://skills-hub.ai/blog/claude-fable-5-coding-workflows-2026): Fable 5 just dropped as Claude Code's new default option — 80.3% on SWE-Bench Pro, the first Mythos-class model cleared for general use, and free on subscriptions through June 22. Here's what changes for your day-to-day coding workflows.+ - [Claude Opus 4.8: 88.6% SWE-bench, Cheaper Fast Mode, and What Developers Need to Know](https://skills-hub.ai/blog/claude-opus-4-8-swe-bench-2026): Opus 4.8 lands at 88.6% SWE-bench Verified and 69.2% on the harder Pro set — nearly 5 points ahead of 4.7. Here's what the benchmark gap means in practice, how the 3× cheaper fast mode changes cost math, and when to route tasks to Opus vs Sonnet.+ - [GPT-5.5 for Agentic Coding: What the Terminal-Bench 2.0 Leader Means for Your Workflow](https://skills-hub.ai/blog/gpt-5-5-agentic-coding-workflows-2026): GPT-5.5 leads Terminal-Bench 2.0 at 82.7% and cuts hallucinations by 52.5% versus GPT-4.5. Here's what changes for developers running agentic coding pipelines — API migration, SKILL.md compatibility, MCP tool-call format, and when to switch.+ - [Claude Code Hooks: Complete Guide to All 30+ Lifecycle Events](https://skills-hub.ai/blog/claude-code-hooks-complete-guide-2026): Claude Code hooks fire deterministically at 30+ lifecycle points — before and after every tool call, on session start and end, when files change on disk, during context compaction. Here's the definitive map with real configs for safety enforcement, auto-formatting, and session context loading.+ - [Hermes Desktop: The MIT-Licensed Agent That Remembers Everything Between Sessions](https://skills-hub.ai/blog/hermes-desktop-persistent-memory-2026): Nous Research shipped a native cross-platform desktop app for Hermes Agent on June 5 — 100+ PRs in a week. Here's what persistent goals, durable kanban, and checkpoint rollbacks mean for developers who run long-horizon work across multiple sessions.
Showing the first 40 of each.