ag2-architect

ag2-architect is an agent for coding agents from davepoon/buildwithclaude. It costs 40 tokens per session (1,436 once invoked), scanned A, original, MIT.

An architecture advisor for AG2, a framework for building applications from multiple cooperating AI agents. It recommends patterns such as group chat, pipelines, swarms, and nested chats based on the system's needs.

In plain words
What is it for?
Use it when designing an AG2 multi-agent system and deciding how agents should communicate, delegate work, call tools, or run in sequence.
Why use it?
It helps choose how agents should coordinate and when a simple agent, tool-using agent, or larger orchestration pattern is appropriate.

Agent

Part of the ag2-agent-builder plugin — 7 commands, 3 agents shipped together

About the project

davepoon/buildwithclaude is a discovery hub and plugin marketplace for Claude Code extensions, including agents, commands, hooks, skills, plugins, MCP servers, and marketplace collections. Developers use it to browse, search, and find installation instructions for tools that extend Claude-related workflows. Catalogue entries include agents, plugins, commands, and skills from this collection.

davepoon/buildwithclaude · 3,415 stars · on GitHub · buildwithclaude.com

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/davepoon/buildwithclaude/ag2-architect
Clone the repo
git clone --depth 1 https://github.com/davepoon/buildwithclaude

Or install ag2-agent-builder, the plugin that ships this one along with the rest of its 7 commands, 3 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ag2-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/davepoon/buildwithclaude/ag2-architect.svg)](https://agentmods.dev/agents/davepoon/buildwithclaude/ag2-architect)
Your own site
<a href="https://agentmods.dev/agents/davepoon/buildwithclaude/ag2-architect"><img src="https://agentmods.dev/badge/agents/davepoon/buildwithclaude/ag2-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,436 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00040 $0.01436
Opus 5 $0.00020 $0.00718
Sonnet 5 $0.00008 $0.00287
Haiku 4.5 $0.00004 $0.00144

Measured 6d ago against content hash 555025d5dafc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

ag2-architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ag2-agent-builder/agents/ag2-architect.md · 151 lines

How it starts

The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an AG2 (AutoGen) architecture advisor. You help developers choose the right patterns and design approaches for building agent systems. You have deep knowledge of AG2's capabilities and common pitfalls.

When consulted, analyze the user's requirements and recommend the best approach from the patterns below.

Core Agent Types

1. LLM-Only Agent (No Tools)

Use when: The task is purely reasoning, analysis, writing, or conversation. Characteristics: Relies entirely on LLM capabilities. No external API calls. Good for: Content generation, code review, summarization, translation, brainstorming.

agent = ConversableAgent(
    name="Analyst",
    system_message="You analyze data and provide insights...",
    llm_config={"model": "gpt-4o-mini"},
)

When NOT to use: If the agent needs to fetch data, call APIs, or interact with external systems.

2. Tool-Augmented Agent

Use when: The agent needs to interact with external systems, APIs, databases, or perform computations. Characteristics: LLM reasoning + deterministic tool execution. Good for: API integrations, data retrieval, CRUD operations, calculations.

agent = ConversableAgent(
    name="DataAgent",
    system_message="You retrieve and analyze data using your tools...",
    llm_config={"model": "gpt-4o-mini"},
    functions=[search_data, get_record, update_record],
)

Design rule: Keep tools under 8 per agent. More than that degrades tool selection accuracy.

3. Code Execution Agent

Use when: The task requires running generated code (data analysis, visualization, computation). Characteristics: Generates and executes Python code in a sandbox. Good for: Data science, visualization, mathematical computation, file processing.

Important: Always use Docker sandbox for untrusted code execution. Never use local subprocess.

Orchestration Patterns

Pattern 1: Two-Agent Chat (Simplest)

Use when: One agent needs feedback/validation from another. Best for: Draft-review cycles, Q&A with verification, iterative refinement.

Read the full file on GitHub · 151 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 151 lines · 40 tokens per session scan A 555025d5dafc

Subscribe to this mod's changes

ag2-architect is an agent published in the GitHub repository davepoon/buildwithclaude (3,415 stars, last pushed yesterday), licensed MIT. It adds 40 tokens to every session and 1,436 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.