agent-architect

agent-architect is a skill for Claude Code from anyaa-labs/agent-skills. It costs 349 tokens per session (21,209 once invoked), scanned A, original, MIT.

A review and design guide for systems where multiple AI agents work together, including their prompts, tools, and surrounding controls.

In plain words
What is it for?
Use it to audit an existing multi-agent system, review a prompt or skill, track findings across sessions, or plan a new agent architecture.
Why use it?
It helps identify weaknesses in an agent setup and provides a structured way to improve or design one.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions CLAUDE.md; mentions subagents; names the AskUserQuestion tool.

Good fit Use it to audit an existing multi-agent system, review a prompt or skill, track findings across sessions, or plan a new agent architecture.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/anyaa-labs/agent-skills/agent-architect
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.

Any agent
npx skills add anyaa-labs/agent-skills --skill agent-architect
Clone the repo
git clone --depth 1 https://github.com/anyaa-labs/agent-skills

Made for: Claude Code.

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 agent-architect

README.md
[![agentmods](https://agentmods.dev/badge/skills/anyaa-labs/agent-skills/agent-architect/github.svg)](https://agentmods.dev/skills/anyaa-labs/agent-skills/agent-architect)
Your own site
<a href="https://agentmods.dev/skills/anyaa-labs/agent-skills/agent-architect"><img src="https://agentmods.dev/badge/skills/anyaa-labs/agent-skills/agent-architect/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-architect

Your own site · 80×15
<a href="https://agentmods.dev/skills/anyaa-labs/agent-skills/agent-architect"><img src="https://agentmods.dev/badge/skills/anyaa-labs/agent-skills/agent-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 349 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 21,209 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00349 $0.21209
Opus 5 $0.00175 $0.10605
Sonnet 5 $0.00070 $0.04242
Haiku 4.5 $0.00035 $0.02121

Measured today against content hash 195f91bc82e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

agent-architect scanned grade A with 1 finding 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 today.

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.

Subtle steeringlowPrompt injection

Instructions that bias recommendations or shape behaviour without the user noticing.

Word the question to match which case actually applies. Never tell the user you have no profile for a model whose profile you are holding.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

agent-architect/SKILL.md · 1,052 lines

How it starts

The opening of the file, as written. The whole thing — 1,052 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Preamble: Update Check

On every invocation, before doing anything else, silently run this command:

SKILL_LINK=""
for candidate in "$HOME/.claude/skills/agent-architect" "$HOME/.agents/skills/agent-architect"; do
  [ -e "$candidate" ] || continue
  SKILL_LINK="$candidate"
  break
done
[ -n "$SKILL_LINK" ] && bash "$(dirname "$(readlink "$SKILL_LINK")")/bin/update-check" 2>/dev/null || true

Interpret the output:

  • If the output contains UPDATE_AVAILABLE: Parse LOCAL_VERSION, REMOTE_VERSION, and the CHANGELOG_DIFF section from the output. Then:

    1. Tell the user a new version is available, showing the version numbers.
    2. Summarize the changelog entries as 3-5 user-facing bullets highlighting new capabilities, improvements, and fixes. Focus on value, not internal details.
    3. Ask the user: "Would you like to update now?" (use AskUserQuestion).
    4. If the user approves: Re-resolve SKILL_LINK using the command above, then run bash "$(dirname "$(readlink "$SKILL_LINK")")/bin/do-upgrade". If output contains UPGRADE_SUCCESS, confirm the new version and continue with the user's original request. If UPGRADE_FAILED, tell the user the auto-upgrade failed and suggest they run git pull manually in the repo directory (shown in the update-check output as REPO_DIR), then proceed with the current version.
    5. If the user declines: Proceed immediately with the current version. Do not mention the update again for the rest of this session.
  • If no output or the command fails: Proceed silently. Never mention the update check to the user.


Platform Tool Mapping

This skill names Claude Code tools because that is the authoring convention. In Codex, use the platform equivalent:

  • Read, Grep, Glob -> native file and shell tools (typically exec_command with rg)
  • Bash -> exec_command
  • Agent -> spawn_agent, wait_agent, close_agent
  • AskUserQuestion -> ask the user directly, or request_user_input when it is available
  • WebSearch -> Codex web search tools

Read the full file on GitHub · 1,052 lines

Files

What ships with it

40 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. today Changed · +102 lines · +38 tokens per session 195f91bc82e8
  2. 9d ago First seen · 950 lines · 311 tokens per session scan A 82fa8fa7d59b

Subscribe to this mod's changes

agent-architect is a skill published in the GitHub repository anyaa-labs/agent-skills (2 stars, last pushed yesterday), licensed MIT. It adds 349 tokens to every session and 21,209 once invoked, about $0.0017 per session on Opus 5. A static security scan graded it A with 1 finding (subtle steering). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens