LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
Pluginofficial publisher

wshobson/agents

Plugin : A set of capabilities installed into an AI application or a host · requires claude-code

Community marketplace of 200+ agents, 180 skills, and 100 commands for Claude Code and others.

Huge and highly starred; pick the useful agents one by one rather than installing everything.

Official site GitHub · 41,653 ★
Overall score
7.8 /10
Security: one caveat, see the card
Utility
8.0
Reliability
10.0
Security
One caveat
Ease
5 min · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Managing dozens of agents, skills, and commands scattered across multiple AI tools requires significant configuration effort. This repository centralizes 94 plugins, 202 agents, 183 skills, and 105 commands into a single source, installable in a few lines within Claude Code, Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, or Pi.

Each plugin is isolated and composable: installing a plugin loads only its components into context, not the entire catalog. Agents cover a wide range of domains including software architecture, security, machine learning, infrastructure, and SEO. A tiered model system automatically routes tasks to Claude Haiku for fast operations, Sonnet for testing and documentation, or Opus for code review and security.

This marketplace is aimed at developers already using Claude Code or another AI command-line tool who want to enrich their environment with specialized agents without configuring them manually one by one. A built-in evaluation framework (plugin-eval) lets you measure the quality of each plugin before adopting it.

This is not a tool for beginners or those looking for a graphical interface: everything happens on the command line, and each activated agent has access to the shell and project files, which requires careful review of what you install.

What it offers

  • ·94 individually installable plugins, each isolated to load only its own agents, skills, and commands into context.
  • ·202 specialized agents covering architecture, programming languages, security, machine learning, infrastructure, documentation, and SEO.
  • ·183 modular skills with progressive disclosure, installable without cloning the repository via gh skill install or npx skills add.
  • ·105 slash commands for code scaffolding, security scans, test generation, and infrastructure setup.
  • ·16 orchestrators to coordinate multiple agents on complex workflows such as full-stack development, incident response, or machine learning.
  • ·Native support for seven environments (Claude Code, Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, Pi) from a single Markdown source.
  • ·Three-layer plugin-eval evaluation framework (static analysis, LLM judge, Monte Carlo) to measure and certify plugin quality.
  • ·HOL Guard integration to scan plugins before activation, with explicit user approval required.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
claude-code, codex, cursor
Host
claude-code
Licence
MIT
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Your data
Data stays local
Maintenance
Active

Links and repository checked on 25 October 2026 · added on 1 August 2025

How to install wshobson/agents

1. Open Claude Code in your terminal. 2. Add the marketplace with /plugin marketplace add wshobson/agents. 3. Browse the catalog in docs/plugins.md to identify plugins useful for your project. 4. Install only the plugins you need: /plugin install <plugin-name>. 5. Before activating a plugin, check its shell and file permissions by reading its content in plugins/<name>/. 6. For other environments (Codex, Cursor, OpenCode, etc.), refer to docs/harnesses.md for the specific procedure.
/plugin marketplace add wshobson/agents

In the same category

Whole category Frame (methods, guardrails) →