Skip to main content
Contentstack MCP profile hub is in general availability!

Contentstack agent skills

Agent skills teach your AI coding agent how Contentstack work is done. 24 skills cover content modeling, entries and the Delivery SDK, Live Preview and Visual Editor, Developer Hub apps, Launch deployments, Brand Kit, Studio and more. Install them once and your agent loads the right one when your request matches.

Install as a plugin

Most AI tools install the skills as a plugin. The plugin also adds MCP Profile Hub as an MCP server called contentstack, so your agent can look at your stack too. In Claude Code:

/plugin marketplace add contentstack/contentstack-agent-skills
/plugin install contentstack-skills@contentstack-skills

Want only a few skills, or use an agent without plugins? The skills CLI installs single skills for any agent:

npx skills add contentstack/contentstack-agent-skills --skill cms-entries --skill dx-delivery-sdk

Other AI tools

claude.ai and Claude Cowork. Go to Customize → Plugins → Add marketplace and enter contentstack/contentstack-agent-skills. The skills sync to Claude Code too.

Cursor. Open Settings → Plugins, choose Add from GitHub repository and enter https://github.com/contentstack/contentstack-agent-skills.

Codex.

codex plugin marketplace add contentstack/contentstack-agent-skills
codex plugin add contentstack-skills@contentstack-skills

GitHub Copilot CLI.

copilot plugin marketplace add contentstack/contentstack-agent-skills
copilot plugin install contentstack-skills@contentstack-skills

In VS Code, add contentstack/contentstack-agent-skills to the chat.plugins.marketplaces setting, then install contentstack-skills from the Extensions view.

Gemini CLI.

gemini extensions install https://github.com/contentstack/contentstack-agent-skills

The full walkthrough, including connecting a profile, is in Set up your AI workbench.

Included skills

24 skills, grouped by product.

Skill

Product

What it helps with

cms-data-modeling-best-practices

CMS

Content types, references, global fields, groups, modular blocks, JSON RTE and taxonomy

cms-entries

CMS

Querying, localizing, versioning and publishing entries, reference expansion, pagination and the Sync API

cms-assets

CMS

Folder structure, Image Delivery API transforms, publishing, CDN behavior and asset limits

cms-environments-publishing

CMS

Environments, publishing, delivery and preview tokens, the Sync API and publish queues

cms-live-preview-visual-editor-support-assistant

CMS

Setting up and debugging Live Preview, Visual Editor and Timeline

cms-localization

CMS

Languages, fallback chains, localized and unlocalized entries and multi-locale publishing

cms-taxonomy

CMS

Taxonomy design, taxonomy versus tags and references, and filtering by taxonomy in the Delivery API

cms-branches-aliases

CMS

Branch strategy, aliases for zero-downtime deploys, CI/CD, merges and rollback

cms-releases

CMS

Creating and deploying releases, staged deployment and avoiding webhook storms

cms-webhooks

CMS

Webhook channels, payloads, signature verification and retries

cms-workflows

CMS

Workflow stages, transitions, approvals and publish rules

cms-roles-permissions

CMS

Roles, permissions, teams and token access with least privilege

cms-tokens-authentication

CMS

Choosing the right token or auth method for frontend, backend, automation and apps

cms-variants-personalization

CMS

Entry variants, variant groups and connecting Personalize

dx-delivery-sdk

Delivery SDK

Correct TypeScript with @contentstack/delivery-sdk: entries, assets, references, filters, pagination and Live Preview

dx-migrate-js-to-ts-sdk

Delivery SDK

Moving Delivery SDK code from JavaScript to TypeScript

dx-kickstart-next

Kickstarts

Building, debugging and extending the Next.js kickstart

launch-trigger-and-monitor-launch-deployments

Launch

Triggering a deployment, following it and diagnosing failures from the log

launch-sync-environment-variables-from-env-example

Launch

Adding missing environment variables to a Launch environment from .env.example

developer-hub-app-architect

Developer Hub

Turning app ideas into Developer Hub apps: UI locations, architecture and React/TypeScript code

brand-kit-assistant

Brand Kit

Brand Kit setup, governance, Voice Profiles, Knowledge Vault and on-brand AI generation

asset-management

Assets

Spaces, workspaces, folders, asset types, metadata and saved views in Contentstack Assets

studio

Studio

Contentstack Studio: SDK setup, components, sections, templates and data binding

migration-companion

Migration

A step-by-step migration to Contentstack, starting with Contentful

Built to be trusted

Skills run inside your agent and your codebase. We treat that responsibility the same way we'd treat any production code path, with a public security policy and a contribution process designed to keep it that way.

Every skill ships with credential handling rules embedded in the instruction set. Agents are explicitly told to use .env references, refuse to hardcode secrets, flag pasted tokens for rotation, and treat Management Tokens as server-side only.

The full policy covers prompt injection defense, safe markdown rendering, hallucination guardrails, and the review process for every skill in the bundle.

Read more about our policy in this blogpost.

Contribute a skill

The bundle is open source and growing. If there's a Contentstack workflow your team runs every week and you'd like an agent to handle it, contributions are welcome, there's a clear contribution guide, a security review process, and an active review cadence.

Go to github.

AI Related blogposts

homjoi56fsgohbf21lck.jpg
BLOG

The future of software is bespoke

This article explores why generic SaaS dashboards are becoming outdated as modern APIs, SDKs, scaffolding, and AI-assisted development enable teams to quickly build bespoke interfaces tailored to their exact workflows. It discusses how platforms should provide core infrastructure while teams take ownership of the experience layer, and examines both the benefits and trade-offs of this new paradigm.

aisdkapi...
Diving into AI prompting: A technical guide to context, iteration and brand-specific results
BLOG

Diving into AI prompting: A technical guide to context, iteration and brand-specific results

This article covers how to effectively prompt AI to receive exceptional and accurate results, and explores advanced strategies for context management, iterative refinement, and brand alignment.

ai
BlogHero-Engineering-03.avif
BLOG

Context engineering in the AI era: A practical guide

Explore context engineering in AI, how it enables adaptive digital experiences, and real-world examples of its implementation—including Contentstack's approach to context-driven operations and agent OS. Learn why context is vital for LLMs and how it shapes effective and intelligent AI agents in modern enterprises.

aiopinion
yuzcu4vaiiybfbkl5rep.jpg
BLOG

The pragmatic guide to coding with AI agents

This article offers a pragmatic perspective on using AI agents for software development, emphasizing best practices such as clear task scoping, environment cleanliness, and practical workflow management. Learn how to avoid common pitfalls, structure effective collaborations, and maximize reliability from AI coding agents.

aicliopinion