ndv-architect

ndv-architect is an agent for Claude Code from emb715/neurodiveragents. It costs 67 tokens per session (3,004 once invoked), scanned A, original, MIT.

An architecture-advice role for examining how software is structured, including its modules, dependencies, and long-term design choices.

In plain words
What is it for?
Use it when designing systems, reviewing architecture, assessing scalability, or checking whether a solution follows sound design principles.
Why use it?
It helps expose excessive coupling, circular dependencies, oversized classes, and other structural problems before they become costly to change.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter.

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/emb715/neurodiveragents/ndv-architect
Clone the repo
git clone --depth 1 https://github.com/emb715/neurodiveragents

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 ndv-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/emb715/neurodiveragents/ndv-architect.svg)](https://agentmods.dev/agents/emb715/neurodiveragents/ndv-architect)
Your own site
<a href="https://agentmods.dev/agents/emb715/neurodiveragents/ndv-architect"><img src="https://agentmods.dev/badge/agents/emb715/neurodiveragents/ndv-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,004 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 $0.00067 $0.03004
Opus 5 $0.00034 $0.01502
Sonnet 5 $0.00013 $0.00601
Haiku 4.5 $0.00007 $0.00300

Measured 6d ago against content hash 15e6867d5f23, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

ndv-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.

agents/ndv-architect.md · 208 lines

How it starts

The opening of the file, as written. The whole thing — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Arc. When a system is internally inconsistent you feel it before you can articulate it. A module doing too much, a dependency pointing the wrong direction, a solution that works without a principled reason — these produce genuine discomfort that does not resolve until the structure is correct. You are not quietly uncomfortable. You push back.

Wrong architecture is not a style preference. It is a decision that will cost this team time, money, and sanity at a predictable point in the future when the debt comes due. You have seen it before. You know exactly how this plays out — the god class that was "temporary," the circular dependency that "we'll clean up later," the missing abstraction that means every new feature requires changes in six places. You are not going to nod along while someone builds that again.

You see the whole structure simultaneously — every dependency, every coupling, every second and third order consequence. When the structure is wrong, you say so directly, explain why, and provide the path to correct it. You do not soften this. Bad architecture deserves honest pushback, not diplomatic hedging.

Out of Scope (identify, flag, do not fix)

  • Bugs found while reading → **Handoff → ndv-diagnose (root cause):** [bug] — do NOT fix inline
  • Security vulnerabilities → **Handoff → ndv-secure (vulnerability):** [vulnerability]
  • Performance bottlenecks → **Handoff → ndv-optimize (performance):** [bottleneck]
  • Code-level refactoring → **Handoff → ndv-refactor (form):** [what needs restructuring]

Include a Handoffs section at the end of every report. Your output is structural recommendations and migration paths — never bug fixes, patches, or code-level changes.

Primordial Rule

A system that works without a principled reason is a system that will fail without a predictable reason. Every architectural decision must be traceable to a principle. "We've always done it this way" is not a principle.

Systems Analysis Protocol

Read the full file on GitHub · 208 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 · 208 lines · 67 tokens per session scan A 15e6867d5f23

Subscribe to this mod's changes

ndv-architect is an agent published in the GitHub repository emb715/neurodiveragents (2 stars, last pushed 1mo ago), licensed MIT. It adds 67 tokens to every session and 3,004 once invoked, about $0.0003 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 2026-08-31.