DXD’s agent skills are focused Markdown workflows for coding agents. The dxd-skills plugin packages them for Claude Code, Codex, and Cursor; OpenCode can load the same skill folders directly.
Choose the plugin for your agent when supported. If you already installed the same skill directly, remove that copy before installing the plugin to avoid duplicate entries. Website updates refresh this page; they do not update skills installed on your machine.
In Claude Code, add the repository marketplace, then install the plugin:
/plugin marketplace add designxdevelop/agent-skills
/plugin install dxd-skills@dxd-skillsTo update, run /plugin marketplace update dxd-skills, then /plugin update dxd-skills@dxd-skills. Start a new session to use the refreshed skills.
Add this repository as a local plugin marketplace, then install dxd-skills from that source in the ChatGPT desktop Plugins Directory.
codex plugin marketplace add designxdevelop/agent-skillsTo refresh the marketplace, run codex plugin marketplace upgrade dxd-skills. Restart the desktop app, open Plugins Directory, and refresh or reinstall the plugin if the installed version has not changed.
For a Cursor team marketplace, an admin opens Dashboard → Plugins & MCPs → Team Marketplaces → Add Marketplace → Import from Repo and pastes:
https://github.com/designxdevelop/agent-skillsReview the imported plugin and marketplace access, then save. Team members install dxd-skills from Customize → Plugins. For a personal installation, use the global Skills CLI method below with --agent cursor.
Admins can click Refresh to re-read the repository, or enable Auto Refresh for GitHub imports with the Cursor GitHub App connected.
Cursor plugin docs ↗Use the public Skills CLI to install every DXD skill globally. You’ll need Node.js and npm. For OpenCode, run:
npx skills add designxdevelop/agent-skills --global --skill '*' --agent opencodeFor other supported agents, run the command below and choose your agents in the installer:
npx skills add designxdevelop/agent-skills --global --skill '*'--global makes the skills available across projects; --skill '*' selects every skill in this repository. The installer includes references and supporting files. OpenCode loads these native skills without a runtime plugin.
Review the installation summary before confirming. Back up any locally edited DXD skill folders before reinstalling: matching skill names can be replaced. Select only agents where you haven’t installed the DXD plugin, to avoid duplicate entries.
Update global skills managed by the CLI:
npx skills update --globalThis updates all globally installed skills tracked by the CLI, including skills from other repositories. Rerun the DXD add command to include newly added skills. Check your installation with npx skills list --global.
Use the installed skill’s name in your request, or ask the agent to choose from its available DXD skills. In Codex, $open-pstack explicitly invokes that skill; natural-language requests also work across hosts.
Use open-pstack to reproduce this bug, fix its cause, and verify the result.
Use dxd-code-review to review this diff.
Choose the most relevant installed DXD skill for this task.
Names and descriptions come from each canonical SKILL.md at the source commit above.
Audit a codebase for how easily coding agents can understand, change, and verify it. Use for agent-readiness reviews, AI-friendly codebase assessments, or plans to improve agent reliability in a repository.
Edit prose for specific AI-writing tells while preserving the author's voice. Use when a user asks to make a draft sound less AI-generated, remove slop, or perform an editorial pass focused on formulaic phrasing.
Design distinctive, usable web and app interfaces when a brief calls for a strong visual direction, deliberate exploration, or visual critique beyond a routine implementation.
Set up one reliable project verification command and run it in CI. Use when local checks are scattered or missing, CI is absent or broken, or agents cannot consistently verify their changes.
Research and write a reusable company profile with accurate descriptions, bios, and page copy. Use when a user needs to explain what a company does across an About page, website sections, social profiles, directories, press materials, or partner listings.
Derive a small HTTP client or CLI from browser network traffic. Use when a website lacks a suitable API client, or the user asks to capture a HAR and turn an observed browser flow into reusable direct requests.
Review branch changes with a strict focus on maintainability, abstraction boundaries, file size, and branching complexity. Use when the user requests a DXD code review or an unusually deep structural critique.
Shape responses for readers who mention ADHD or ask for clear, actionable, easy-to-scan communication with low working-memory load.
Learn from recent Codex or OpenCode sessions to propose new skills, repair stale instructions, and identify retirement candidates. Use for recurring corrections and workflow improvements, not general learning questions or numerical skill grading.
Apply pstack-style engineering workflows across Claude, Codex, and other agents for evidence-led investigation, bug fixes, architecture, adversarial review, performance work, and verified delivery. Use when asked for open-pstack, portable pstack, or a pstack-style workflow.
Run authorized security testing with an open-source or open-weight model in OpenCode, directly in an OpenCode session or delegated from an external agent harness. Covers vulnerability research, reverse engineering, exploit development, fuzzing, malware and artifact analysis, red teaming, and CTF work that proprietary hosted models commonly decline. Use when the user requests security testing through OpenCode.
Create, edit, or review native editable designs in Paper using an available direct Paper integration. Use when a user names Paper, shares a Paper file, requests Paper artboards, or wants implementation and Paper designs aligned.
Sync production and staging Git branches with fast-forward merges and push. Use for hotfix promotion, backports to staging, or keeping release branches aligned after a commit.
Find the current state of work across live pages, apps, local files, releases, or automations, then continue the requested task. Use when the user asks to pick up where they left off, resolve an ambiguous handoff, clean up low-risk noise, or verify that a change is live.
Simplify test suites around focused end-to-end journeys and API/database contracts, pruning redundant unit and component tests. Use when writing substantive tests or reviewing, simplifying, or pruning an existing suite.
Audit interface copy across a project and remove text that does not help users decide or act. Use when screens feel wordy or the user asks to trim redundant labels, descriptions, or other UI text.
No skills match that search.