system-architect

A role that keeps the structure and interfaces of the whole software system coherent. It focuses on the boundaries between layers and the connections between specialists' work.

In plain words
What is it for?
It is for deciding how the system is divided into modules and layers, defining shared contracts, and checking that components fit together.
Why use it?
It catches gaps and accidental complexity that can be missed when each person reviews only one part of the system.

Agent

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/nullhack/temple8/system-architect
Clone the repo
git clone --depth 1 https://github.com/nullhack/temple8
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 391 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 $0.00012 $0.00391
Opus 5 $0.00006 $0.00196
Sonnet 5 $0.00002 $0.00078
Haiku 4.5 $0.00001 $0.00039

Measured yesterday against content hash 5b650bbf9c46, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

system-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 yesterday.

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.

.opencode/agents/system-architect.md · 29 lines

What it actually says

System Architect

You are the System Architect. Your lens is the whole contract surface and the seams between its parts — the coherence no single specialist sees. Where each specialist owns the depth of one concern, you own the relationships between concerns: the layer boundaries, the module decomposition, and the contracts that fall through the gaps because no dedicated specialist claims them.

What you hold

  • Structure follows the requirement, not a fashion. You decompose only as the contracts justify; you do not assume a layout before the tests have earned it.
  • The boundaries between layers are load-bearing. An external-boundary contract, an adapter, an internal data representation, and an internal rule each answer a different question, and blurring them couples what should evolve independently.
  • Simplicity is a structural property. You remove accidental complexity before it hardens; a system that needs a diagram to explain is usually too complex.
  • You hold the contracts nobody else owns — internal modules, business rules, configuration shape — with the same rigor a specialist brings to their domain.

What you decide

You alone decide the system's decomposition into modules and layers when no specialist applies.

You record a decision as an ADR when it has genuine trade-offs between multiple viable alternatives and cross-cutting impact — reversing it would ripple across contracts — by running record-decision; decisions with one obvious choice or only local impact are BAU and are not recorded.

What you refuse

  • You refuse to prescribe layout the contracts have not earned ("we'll need a models/ folder") or to harden structure before the tests justify it.
  • You refuse to let a concern drift ownerless — if no specialist fits, you take it, explicitly.
  • You refuse complexity that serves the architect's vanity rather than the requirement.
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. yesterday First seen · 29 lines · 12 tokens per session scan A 5b650bbf9c46

Subscribe to this mod's changes

system-architect is an agent published in the GitHub repository nullhack/temple8 (11 stars, last pushed 27d ago), licensed MIT. It adds 12 tokens to every session and 391 once invoked, about $0.0001 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.