Plugins, agents, and skills for AI coding tools

Use one collection of plugins, agents, skills, and commands across Claude Code, OpenAI Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, and Pi. A harness is the coding tool that loads and runs the components. Plugins cover Python and JavaScript development, code review, testing, infrastructure, security, and other work.

The catalog contains 94 plugins, including 92 local plugins and 2 external entries. The local source contains 202 agents, 184 skills, and 105 commands. The components share a Markdown source, with installation and capabilities adapted to each harness.

Multi-harness support

Choose the harness you use and follow its setup route. Available components depend on the installation route and the harness's capabilities.

Tool	Installation	Components and limits
Claude Code	Plugin marketplace	Source agents, skills, and commands; plugin-specific hooks and MCP configuration
OpenAI Codex CLI	Native marketplace or generated artifacts	Native source skills; generated TOML agents and commands as skills use a separate route
Cursor	Plugin marketplace	Source agents, skills, and commands, with curated project rules
OpenCode	make install-opencode	Generated agents, skills, and commands, with translated tool permissions
Antigravity CLI	make install-antigravity	Generated plugins containing agents, skills, and commands
GitHub Copilot	make install-copilot	Generated agent profiles and skills; commands also become invocable skills
Pi	make install-pi	Generated skills and prompt templates; agents require a subagent extension

See the capability matrix for model mappings, tool permissions, hooks, and installation differences.

For individual skills, use a skills-only installer and select your coding tool.

Quick start

Choose a plugin for the work you do:

Plugin	Use it for
python-development	Python, Django, FastAPI, testing, and packaging
javascript-typescript	JavaScript and TypeScript development and project setup
developer-essentials	Code review, debugging, Git, and testing patterns
security-scanning	Security review, dependency checks, and code scanning

See the plugin catalog for all 94 plugins and the usage guide for commands and examples.

Claude Code

Run these commands inside Claude Code to register the marketplace and install the Python development plugin:

/plugin marketplace add wshobson/agents
/plugin install python-development@claude-code-workflows

Then ask Claude to use the Python agents for your task, or run the plugin's scaffolding command:

/python-development:python-scaffold Create a FastAPI service with tests

Codex CLI

Run these commands in a terminal with a current Codex CLI:

codex plugin marketplace add wshobson/agents
codex plugin add python-development@claude-code-workflows

The native Codex manifests expose source skills. The generated TOML agents and command-derived skills use a separate adapter path. Native installation also reads the original skill bodies, while generated copies split oversized bodies into reference files for the 8 KB budget. See the Codex setup details before choosing an installation route.

Cursor

Cursor uses the committed plugin marketplace registry. Follow the Cursor setup instructions to add the marketplace and install a plugin.

Skills only

Install an individual skill with GitHub CLI's gh skill command or the npx skills installer. Select your coding tool when prompted. Both installers read skills from GitHub without cloning the repository and install them without the plugin's agents, commands, or hooks.

gh skill install wshobson/agents python-testing-patterns
npx skills add wshobson/agents --skill python-testing-patterns

gh skill requires GitHub CLI 2.90 or later. The npx command requires Node.js and npm. See skills-only installation for other agents, install scopes, and version pinning.

Install from a clone

For OpenCode, Antigravity CLI, GitHub Copilot, and Pi, clone the repository and use the installer for your harness. Install uv, Python 3.12 or later, and the coding tool you want to use. The following clone command uses GitHub CLI:

gh repo clone wshobson/agents ~/agents
cd ~/agents

Run the command for your tool. Each target generates the artifacts and links all local plugins into that tool's user configuration:

make install-opencode
make install-antigravity
make install-copilot
make install-pi

Pi's agents require its reference subagent extension or a compatible extension. See the global install guide for configuration paths and uninstall commands.

What's inside

Component	Count	Purpose
Plugins	94	Installable groups of components, including 92 local plugins and 2 external entries
Agents	202	Instructions for specialist subagents that handle delegated work
Skills	184	Guidance and reference material that an agent loads when relevant
Commands	105	Named workflows you invoke, such as project scaffolding or a security scan

The agent, skill, and command counts cover local source files. External plugins provide their own components and installation requirements.

How it works

A plugin groups related agents, skills, and commands under plugins/. For example, python-development contains three agents, sixteen skills, and a scaffolding command. The chosen harness loads the installed components according to its own discovery and invocation rules.

Contributors maintain one Markdown source in the Claude Code plugin format. Native registries point to that source where supported, and adapters generate the formats needed by each harness. Commands may become skills or prompt templates, and agent formats, model choices, and tool permissions vary by harness.

Source agents specify model aliases such as opus, sonnet, haiku, fable, or inherit. Adapters map those aliases to each harness's model settings, including fixed defaults for some inherit entries. See model configuration and the adapter mappings.

Quality evaluation

The repository validates plugin structure, generated artifacts, and documentation. Contributors can run these checks from the repository root:

make generate-all
make validate STRICT=1
make garden

plugin-eval also provides static skill checks without model calls. For example:

uv run --project plugins/plugin-eval plugin-eval score plugins/python-development/skills/python-testing-patterns --depth quick

The LLM judge and Monte Carlo layers are experimental and are not validated against human labels. Their scores describe model assessments and repeated simulations and do not prove that a task succeeds in a project. They require the optional llm dependencies. Model calls use ANTHROPIC_API_KEY when it is set and bill that API account. Otherwise they use your Claude Code login. See the evaluation guide for setup and limits.

Documentation map

Guide	Covers
Plugin catalog	Plugins by task and domain
Agents and skills	Available components and model configuration
Usage and tool setup	Commands, examples, and installation
Architecture and design details	Source layout and adapters
Quality evaluation and CLI verification	Checks and their limits

See CONTRIBUTING.md to propose a change and authoring conventions to write plugin content.

External memory integration

Pensyve remains an optional external Claude Code marketplace entry. Major7 Apps maintains Pensyve, and this repository's maintainer founded Major7 Apps. Its open-source runtime can run locally or on your own server. Pensyve Cloud closed on October 1, 2026, and its dashboard, API, and hosted MCP endpoints are unavailable. Use the self-hosting guide when configuring the runtime.

License

The repository is MIT licensed. External plugins have their own licenses.

Star history

---

TalentDelta provenance record (added at packaging)

• Packaged for: TalentDelta catalogue (draft; gated by review)
• Upstream: https://github.com/wshobson/agents
• Pinned commit: 46891e7e60da0e52baf1050b7b6391b64e84c6d9 (2026-10-04)
• Licence: MIT - the upstream LICENSE (MIT, Seth Hobson) ships unchanged in this pack, and each plugin manifest declares its own license
• Upstream author credit: Seth Hobson (wshobson/agents)
• Runtime: Claude Code, OpenAI Codex CLI, Cursor, OpenCode, GitHub Copilot CLI, Antigravity CLI and Pi - the pack ships a committed plugin manifest for each harness (.claude-plugin/, .codex-plugin/, .cursor-plugin/), so it installs natively on any of them
• Changes made at packaging: this section was appended to README.md. No upstream path is excluded. No file was added, removed, renamed or edited; every file is byte-identical to the pinned commit.
• Safety read of the packaged archive (packaging time, 2026-10-08): scanned before packaging (leak_scan.py over 1134 text files): no live credentials, no personal data, no host paths. All 493 matches were read: 487 are email-shaped strings, of which 155 are the plugin-author metadata that upstream publishes in its harness marketplace indexes (.claude-plugin/, .cursor-plugin/) and the rest illustrative example.com/example.org addresses inside skill docs and test fixtures; the 5 api-key-shaped hits are fake tokens in tools/tests/test_cli_smoke.py (literally "sk-ant...test") plus one already-redacted export line in docs/round-trip-results.md; the single host path is the test fixture C:/Users/smoke. Every shipped byte is upstream's own file at the pin.

Packaging notes (TalentDelta)

• Scouting basis: owner asked for the best / most comprehensive / best-known and best-liked coding agent pack that ideally works with Codex or Claude Code, or other best-known coding-specific agents.
• Why this pack wins: it is the only candidate that is simultaneously (a) a large roster of full agent definitions - 202 agent files, each with YAML frontmatter (name, description, tools, model) and mission/behaviour sections, packaged per plugin with its own manifest and README, (b) natively multi-harness with Codex CLI and Claude Code support committed in-tree rather than bolted on, and (c) actively maintained at 40,301 stars / 4,290 forks with the last commit on the pin date.
• Runner-ups, with the evidence that ruled them out: obra/superpowers (296,634 stars, MIT) is the most-liked candidate but ships 74 skills and zero agent definitions - a skill bundle, not an agent pack; bmad-code-org/BMAD-METHOD (53,934 stars, MIT, active releases) restructured in v6 to 435 skill files, so it is now a methodology/skill system; github/spec-kit (140,648 stars, MIT) is a spec-driven development toolkit; VoltAgent/awesome-claude-code-subagents (25,584 stars, MIT) is a genuine 100+ subagent roster but single-harness with lighter docs; davepoon/buildwithclaude, SuperClaude_Framework (v4.3.0, Mar 2026) and ruvnet/claude-flow are large collections or orchestration frameworks with thinner agent documentation.
• Maturity signals: 40,301 stars, 4,290 forks, created 2025-07-24, last commit 2026-10-04 (the pin), 1150 tracked files / 7.85 MB, MIT LICENSE at the root, per-plugin MIT manifests, 10 shipped test/adapter scripts, and its own docs set (docs/agents.md, docs/harnesses.md, docs/plugin-eval.md, ARCHITECTURE.md). No tagged releases at the pin - the commit hash is the anchor.
• What a busy user gets: a complete coding-agent department - agent-teams (team-lead, implementer, reviewer, debugger) for parallel work with file ownership, plus backend, frontend/mobile, API, database, testing, review, refactoring, cleanup, debugging, security, CI/CD, deployment, observability and documentation plugins, all installable into the harness they already use.