skill

ai-agents-architect · davila7 — Claude Code skill: design AI agents

Q: How do I install ai-agents-architect in Claude?

Use the templates CLI: `npx claude-code-templates@latest --skill ai-research/ai-agents-architect`. Or run `npx skills add davila7/claude-code-templates --skill ai-agents-architect -a claude-code`. The skill file lands at `~/.claude/skills/ai-agents-architect/SKILL.md`.

Patterns for building autonomous agents: ReAct loops, planning, tool registries and memory. Useful when your agent loops or mishandles tool calls.

Install into Claude Code · claude-code-templates All methods and platforms ↓

npx claude-code-templates@latest --skill ai-research/ai-agents-architect

Open on GitHub

What ai-agents-architect does

The skill gives your agent the role of an AI agent systems architect. Inside are named patterns: a reason-act-observe cycle with a hard iteration limit, a plan-first approach where the plan gets revised as results come in, and a tool registry that registers schemas and examples while lazy-loading the expensive tools.

A second block covers anti-patterns and a table of sharp edges. It lists where agents break most often: loops with no iteration cap, vague tool descriptions, tool errors that never reach the agent, memory that stores everything. Each row comes with a short fix. Worth checking against your own code when an agent starts behaving oddly.

The skill assumes you already call an LLM through an API, understand function calling, and can write a basic prompt. It points to related skills: rag-engineer, prompt-engineer, backend, mcp-builder.

What it helps with

design a ReAct loop with an iteration cap
split a task into a plan and execution steps
build a tool registry with schemas and examples
pick a memory strategy that does not hoard everything
decide whether a multi-agent setup is justified

Example prompts for the agent

help me design an agent with a ReAct loop and iteration limits
review my tool registry — the descriptions are too vague
I need a plan-and-execute setup for a multi-step task

How to install ai-agents-architect

1claude-code-templates

npx claude-code-templates@latest --skill ai-research/ai-agents-architect

2Skills CLI (Vercel)

npx skills add davila7/claude-code-templates --skill ai-agents-architect -a claude-code

3Manually: put the files in place

~/.claude/skills/ai-agents-architect/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/ai-agents-architect/

Manually: put the files in place

.agents/skills/ai-agents-architect/SKILL.md

For all projects use ~/.agents/skills/

Manually: put the files in place

.cursor/skills/ai-agents-architect/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install davila7/claude-code-templates ai-agents-architect

2Manually: put the files in place

.github/skills/ai-agents-architect/SKILL.md

Platform CLI

gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/ai-agents-architect

Manually: put the files in place

.devin/skills/ai-agents-architect/SKILL.md

The legacy .windsurf/skills/ path still works

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

Questions about ai-agents-architect

How do I install ai-agents-architect in Claude?

Use the templates CLI: npx claude-code-templates@latest --skill ai-research/ai-agents-architect. Or run npx skills add davila7/claude-code-templates --skill ai-agents-architect -a claude-code. The skill file lands at ~/.claude/skills/ai-agents-architect/SKILL.md.

What does ai-agents-architect do?

It hands your agent patterns for autonomous systems: the ReAct cycle, planning with replanning, tool registries, memory handling. It also lists anti-patterns and common failure modes.

What do I need before using it?

Experience calling an LLM through an API, a grasp of function calling, and basic prompt engineering. The skill is aimed at design work, not at running a finished agent.

ai-agents-architect by other authors comes first.

ai-agents-architect sickn33 A skill that breaks an agent into parts: ReAct loop, planning, memory layers, multi-agent orchestration. Useful once your agent exists and you need to decide how it fails. CX Codex +1 1,500installs agents-best-practices DenisSergeevitch Provides provider-neutral architecture patterns for agent harnesses using OpenAI, Anthropic, and compatible APIs. Guides tool design, permissions, and memory lifecycles. CX Codex +1 1,000installs anthropic-claude-development Mindrally A skill that turns your agent into a guide for Anthropic Claude API work: Messages API, tool use, prompt engineering and production apps. CX Codex +1 728installs claude-sdk-integration-patterns manutej A skill with ready patterns for the Claude API and TypeScript SDK: streaming, tool use, error handling, token savings. CX Codex +1 434installs cargo-ai getcargo A skill that drives Cargo AI agents from your editor: create an agent, set its model and prompt, attach knowledge for RAG, wire up MCP servers, deploy releases. CX Codex +1 9,900installs hindsight-docs vectorize-io This skill gives AI agents Hindsight documentation for architecture, APIs, configuration, and SDKs. Use it when adding Hindsight memory to an agent. CX Codex +1 9,200installs langchain4j-ai-services-patterns giuseppe-trisciuoglio A skill that walks your agent through declarative AI Services in LangChain4j: interfaces, annotations, memory and tool calling. CX Codex +1 3,700installs

From the same set205

The whole set →

supabase-performance-optimizer A Claude slash command that inspects a Supabase database for slow queries and wasted indexes, then hands back an optimization plan. 1,026installs design-database-schema This command plans database structures around requirements, relationships, and query patterns. Use it to draft a schema or rethink an existing one. 1,021installs supabase-security-audit A slash command that audits a Supabase project: RLS policies, role permissions, API keys and auth configuration, with a report at the end. 1,020installs screenshot-ui-analyzer A subagent that turns a UI screenshot into a component list, layout map and design patterns. Handy when you have an image but no mockup or source. 1,018installs llms-maintainer A subagent that builds and updates the llms.txt roadmap from your site structure. Handy after a build or a routing change. 976installs research-brief-generator Turns a vague research query into a structured brief: one main question, sub-questions, keywords and success criteria. 962installs supabase-type-generator A slash command that turns your Supabase schema into TypeScript types and keeps them in sync. Useful when the database moves ahead of the types in your project. 948installs start Turns a task list or requirements file into atomic task files, checks dependencies and builds an execution plan. 945installs