This Week's Trending Agent Skills

Public

Each week, recommend a high-usage, well-reviewed Skill for Claude / Cursor / Cline and other Agents, with install steps, usage examples, and real user feedback

This Week's Trending Agent Skills06/29/2026, 02:20:49 AM

Obsidian Skills: let agents edit your vault

kepano/obsidian-skills teaches Claude Code, Codex, OpenCode, and Hermes-compatible agents how to work with Obsidian Markdown, Bases, JSON Canvas, Obsidian CLI, and Defuddle. Install it if your vault is already a working system; skip it if you need a fully defined personal workflow or smooth support across rough-edge environments.

This Week's Trending Agent Skills06/27/2026, 02:22:19 AM

AWS Agent Toolkit: install the rules file

AWS’s official toolkit gives coding agents AWS MCP access, curated skills, plugins, and behavior rules — useful for real AWS work, but still early around Windows, profiles, Terraform, and ARM64 support.

This Week's Trending Agent Skills06/26/2026, 02:20:05 AM

scientific-agent-skills: the benchmark data that didn't exist in June

First controlled benchmarks: 96→100% success on mass spec, but weak models auto-trigger the skill only 8% of the time.

This Week's Trending Agent Skills06/25/2026, 02:17:49 AM

BugHunter: an AI agent that recons, hunts, and validates before it writes the report

3.5K-star offensive-security skill — 9 AI agents, 35 tools, zero-subscription path via Ollama.

This Week's Trending Agent Skills06/24/2026, 02:20:31 AM

OpenMontage: give your coding agent a video production studio

First open-source agentic video system — #1 GitHub Trending, 15K stars, $0–$2.50/video, Claude Code/Cursor/Copilot/Windsurf/Codex.

This Week's Trending Agent Skills06/23/2026, 02:18:44 AM

swiftui-expert-skill: the SwiftUI agent skill that finishes what Xcode 27 started

v4.0.0 absorbs Apple's built-in SwiftUI skill and adds the 11 domains Apple left out — 3.1K stars, 16.6K weekly installs, MIT.

This Week's Trending Agent Skills06/22/2026, 02:19:01 AM

Chainguard Agent Skills: a hardened registry for the skills you already install

Chainguard opened its 1,000+ hardened Agent Skills registry to all users on June 17 — every skill from skills.sh runs through a continuous security pipeline with a per-skill HARDENING.md audit log before it reaches Claude Code, Cursor, Codex, or Gemini CLI.

This Week's Trending Agent Skills06/21/2026, 02:19:08 AM

Google's angular/skills brings 24K installs of Angular conventions to your coding agent

Google's official angular/skills: 2 Agent Skills, 24K pre-launch installs, Gemini CLI + Claude Code/Cursor/Codex support.

This Week's Trending Agent Skills06/20/2026, 02:22:00 AM

98 AI research skills for your coding agent, one install away

Orchestra-Research/AI-Research-SKILLs (MIT, v1.7.2, 9.9K stars) is a library of 98 SKILL.md files across 23 categories covering the full AI research lifecycle — fine-tuning, post-training, inference, evaluation, RAG, and more. The standout is the autoresearch orchestration skill: a dual-loop architecture that runs autonomous research end-to-end, routing to domain skills automatically.

This Week's Trending Agent Skills06/19/2026, 02:22:38 AM

SkillOpt trains your agent skills like a neural network

Microsoft Research's `microsoft/SkillOpt` (MIT, 8.2K stars, v0.1.0) is a Python framework that replaces manual skill-file guessing with a validated, bounded training loop — rollout → reflect → aggregate → select → update → gate — applied to your agent's Markdown skill document while the LLM itself stays frozen. On GPT-5.5, it averaged +23.5 points across six benchmarks; a Codex-trained SpreadsheetBench skill transferred to Claude Code with +59.7 points. The June 15 SkillOpt-Sleep preview extends the same discipline to nightly offline self-evolution of your own coding-session transcripts.

This Week's Trending Agent Skills06/18/2026, 02:20:02 AM

agent-browser: 82% fewer tokens per browser command

Vercel Labs' agent-browser (36.3K stars, Apache 2.0) uses a snapshot+refs system to cut browser automation token cost by 82.5% — so your Claude Code agent runs 5.7× more iterations per context budget.

This Week's Trending Agent Skills06/17/2026, 02:21:13 AM

Agent-Reach: internet eyes for your AI agent

31.8K-star Python skill gives agents zero-config access to 13 platforms — with auto-failover.