What problem it solves

Coding tools use different component formats, model names, permissions, and discovery rules. Contributors need a shared way to write reusable plugins, agents, skills, and commands once while producing compatible artifacts for each supported harness.

Key features

• Multi-harness component support — One collection of plugins, agents, skills, and commands can be used across Claude Code, OpenAI Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, and Pi.

  Source: README, docs

• Markdown-first source authoring — Contributors maintain each component once in Markdown using the Claude Code plugin format. Adapters then generate the formats required by each harness.

  Source: README

• Automated generation and validation — CI regenerates artifacts for every harness, validates them, and runs selected artifacts through their installed CLIs. Contributors can also regenerate, validate, and maintain the repository with the provided Make targets.

  Source: README, Makefile

• Broad plugin and component catalog — The catalog covers Python and JavaScript development, code review, testing, infrastructure, security, and other work. Its local source includes agents for specialist work, skills for guidance, and commands for named workflows.

  Source: README

• Quality evaluation tooling — Static skill checks are available without model calls, while experimental evaluation layers provide model assessments and repeated simulations. The README states that these scores do not prove that a task succeeds in a project.

  Source: README, evals

Quick start

Copied verbatim from the project's README.

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

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

Who it may suit

This collection may suit contributors and coding-tool users who want to maintain reusable Markdown components across multiple AI coding tools.

Topics to discuss

• Which harness-specific format differences should adapters handle next?

• How should contributors evaluate skills without treating model-based scores as proof of task success?

• Which additional workflows or domains should the plugin catalog cover?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/wshobson/agents#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.