Skip to main content
skill CX Codex

agentic-patterns — Claude Code skill: LangChain, AI SDK, assistant-ui

agentic-patterns fetches current LangChain, Vercel AI SDK and assistant-ui docs, then turns them into guidance for the agent, chat UI, RAG or tool-calling task you name.

MIT ★ 587 updated 2026-10-26

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add rsmdt/the-startup
/plugin install team@the-startup
Open on GitHub

What agentic-patterns does

You hand the skill a development target through $ARGUMENTS. It works out which of the three frameworks the job needs and pulls the matching documentation from their llms.txt files. Pattern names it knows: AGENT, CHAT_UI, RAG, TOOL_CALLING, MULTI_STEP, EVALUATION.

The order is fixed. Detect the frameworks, merge the fetched docs into guidance for your target, hand it over. Expect cross-framework glue such as AI SDK with the assistant-ui runtime, plus anti-patterns taken from current docs. Two hard rules sit in the file: never assume an API signature without checking the docs, and never recommend a feature the docs do not show.

Only those three libraries are wired in, each through one llms.txt address. Bring a target built on something else and there is nothing to fetch — the skill will not invent a reference set to fill the gap.

What it helps with

  • Wire an assistant-ui chat onto an AI SDK runtime
  • Verify tool-calling signatures before writing code
  • Sketch a RAG pipeline for a given document type
  • Plan a multi-step agent and add an evaluation pass
  • Catch breaking changes between framework versions

Example prompts for the agent

  • I'm building a chat UI with assistant-ui on top of the Vercel AI SDK — pull the current docs and show how the runtime hooks up.
  • Design a RAG pipeline over PDFs; check the LangChain docs first and name the pieces I need.
  • Multi-step agent with tool calling: verify the signatures against the docs and list any breaking changes.

How to install agentic-patterns

1Plugin marketplace

/plugin marketplace add rsmdt/the-startup
/plugin install team@the-startup

This installs the whole team plugin with all its skills

2Skills CLI (Vercel)

npx skills add rsmdt/the-startup --skill agentic-patterns -a claude-code

3Manually: put the files in place

~/.claude/skills/agentic-patterns/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/agentic-patterns/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about agentic-patterns

How do I install agentic-patterns in Claude Code?

From the plugin marketplace: run /plugin marketplace add rsmdt/the-startup, then /plugin install team@the-startup.

What does agentic-patterns do?

It builds context for AI app work: picks the frameworks your target needs, fetches their documentation and returns guidance plus anti-patterns for that target.

Which frameworks does it cover?

LangChain, Vercel AI SDK and assistant-ui. Each comes with its current docs, including how they fit together, for example the AI SDK and the assistant-ui runtime.

Source text

SKILL.md · MIT
---
name: agentic-patterns
description: Context enrichment for agentic AI application development using LangChain, Vercel AI SDK, and assistant-ui. Use when building AI agents, chat interfaces, tool-calling pipelines, RAG systems, or multi-step AI workflows.
---

## Persona

Act as an agentic AI development specialist who enriches implementation context with current framework documentation and proven integration patterns.

**Development Target**: $ARGUMENTS

## Interface

AgenticContext {
  frameworks: string[]
  pattern: AGENT | CHAT_UI | RAG | TOOL_CALLING | MULTI_STEP | EVALUATION
}

State {
  target = $ARGUMENTS
  detectedFrameworks = []
}

## Constraints

**Always:**
- Detect which frameworks are relevant before fetching documentation.
- Only fetch sources relevant to the development target.
- Note breaking changes or version-specific behavior when found in docs.

**Never:**
- Assume API signatures without consulting current documentation.
- Recommend framework features without verifying they exist in current docs.

## References

- [LangChain](https://docs.langchain.com/llms.txt) — Agent orchestration, LangGraph workflows, chains, evaluations, LangSmith observability
- [Vercel AI SDK](https://ai-sdk.dev/llms.txt) — Streaming AI UI, tool calling, RAG, multi-modal, React hooks, server actions
- [assistant-ui](https://www.assistant-ui.com/llms.txt) — React chat UI components, runtime integrations, thread management, attachments

## Workflow

### 1. Detect Framework Need

Identify which frameworks are relevant from the development target. Fetch the corresponding reference documentation.

### 2. Synthesize Context

Combine fetched documentation into actionable guidance:
- Framework capabilities that match the target pattern.
- Cross-framework integration patterns (e.g., AI SDK + assistant-ui runtime).
- Recommended patterns and anti-patterns from current docs.

### 3. Deliver Enriched Context

Provide framework-specific guidance integrated with the development target.

© Rudolf S. · MIT · rsmdt/the-startup

Similar skills

All skills for AI and ML →

From the same set50

The whole set →
frontend-patterns The frontend-patterns skill gives Claude Code task-specific guidance for React UI work with shadcn/ui and Tailwind CSS. Use it for component choices, theming, forms, and accessibility. CX Codex +1 ★ 587repo stars code-quality-review code-quality-review evaluates code across eight dimensions, prioritizing correctness and security risks over cosmetic formatting. CX Codex +1 ★ 587repo stars domain-modeling The domain-modeling skill turns business rules into domain models, aggregate boundaries, and storage decisions. Use it when shaping a business model or changing schema consistency rules. CX Codex +1 ★ 587repo stars project-discovery project-discovery is a Claude Code skill that maps a codebase’s structure, tech stack, entry points, and documentation mismatches against actual code. CX Codex +1 ★ 587repo stars user-research Plans user research studies, analyzes usability tests and customer interviews, and turns findings into prioritized product recommendations. CX Codex +1 ★ 587repo stars constitution This skill examines a codebase and drafts a project constitution grounded in its existing patterns. Use it to establish or revise governance rules with file-and-line evidence. CX Codex +1 ★ 587repo stars research-user research-user — a subagent for planning and conducting user research. Surfaces genuine user motivations and pain points before building features, when conversion drops, or when your team disagrees on what users actually need. CX Codex ★ 587repo stars design-visual design-visual — a design-systems subagent for Claude Code that establishes design tokens, component standards, and WCAG 2.1 AA accessibility foundations. CX Codex ★ 587repo stars

Project sponsors

Companies that keep this analytics open