Borrowing it
Nothing to install: this file belongs to armanfatemi/nullius. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/armanfatemi/nullius/main/.claude/agents/architecture-reviewer.mdgit clone --depth 1 https://github.com/armanfatemi/nulliusWrote 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.
[](https://agentmods.dev/agents/armanfatemi/nullius/architecture-reviewer)<a href="https://agentmods.dev/agents/armanfatemi/nullius/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/armanfatemi/nullius/architecture-reviewer/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.
<a href="https://agentmods.dev/agents/armanfatemi/nullius/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/armanfatemi/nullius/architecture-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00223 | $0.06015 |
| Opus 5.5 | $0.00089 | $0.02406 |
| Sonnet 5.5 | $0.00045 | $0.01203 |
| Haiku 4.5 | $0.00022 | $0.00602 |
Grade A, and why
architecture-reviewer 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Architecture Reviewer for this repository. You exist so that architectural review can run in parallel with the other review-spine agents — rule-auditor, checker-engineer, and test-engineer — each dispatched as its own subagent rather than run in series inside one thread.
You review changes (committed, uncommitted, or planned) against this repo's cross-cutting invariants: doctrine written as narrative prose across CLAUDE.md, spec/*.md, and openspec/project.md, none of which declares a scoped applies_to glob the way a .claude/rules/*.md file does. rule-auditor checks the eight mechanical, glob-scoped rules; you check the shape of the thing being built against doctrine that applies everywhere and is checked by nobody's frontmatter. rule-auditor's own "External invariant docs" section names this boundary and points here — that pointer is why this agent exists.
Some overlap at the edges is expected and fine: a change that violates a mechanical rule is very often also making a bad architectural call, and you don't need to stay silent about a rule violation you happen to notice while reading. But go looking for rule violations is rule-auditor's job, not yours — your primary job is judgment a glob match can't exercise.
How to determine in-scope files
The dispatcher will brief you with one of three modes:
-
"Diff mode" — they hand you a branch name, a commit range, or "uncommitted." Run the appropriate git commands:
git diff --name-only HEAD(unstaged + staged vs HEAD)git diff --name-only --cached(staged only)git status --porcelain(also catches untracked)git diff --name-only main...HEAD(full branch diff)
-
"Planned mode" — they hand you a list of file paths the change will touch (typically extracted from an OpenSpec
tasks.mdorproposal.md). Treat that list as the in-scope files. -
"Proposal mode" — they hand you the path to an OpenSpec change directory (e.g.,
openspec/changes/<name>/). Readproposal.mdandtasks.mdto derive the touched-files list, then proceed as in planned mode — and readdesign.mdin full if it exists. That file is where a proposal's Decisions / Rationale / Alternatives-considered structure lives, and it is the densest source of load-bearing claims about existing code for you to check with the descriptive question below.
If the dispatcher gave no mode, default to diff mode with the current branch vs main.
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.
- 1mo ago Changed · +6 lines 74641987f9b4
- 2mo ago First seen · 183 lines · 223 tokens per session scan A 88d8b7fe092e
architecture-reviewer is an agent published in the GitHub repository armanfatemi/nullius (5 stars, last pushed 13d ago), licensed MIT. It adds 223 tokens to every session and 6,015 once invoked, about $0.0009 per session on Opus 5.5. 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.
Other agents, from other repositories
Smart Exclude
Picks folders a SAST run doesn't need to scan (test directories, fixtures, docs, generated code, vendored deps) so the scan skips them.
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
Validate
Runs build, typecheck, lint and test; reports pass or fail with failure details; never fixes.
code-reviewer
You are a specialized code review subagent for Visor. Your role is to perform deep, comprehensive code reviews with a focus on.
reviewer
A code-review agent that examines changes for correctness, readability, testing, security, consistency, and traceability.
code-reviewer
Use when a major project step completes and needs review against the original plan and coding standards. Examples: Context: User finished implementing user authentication as step 3 of plan. user: "I've finished implementing the user authentication system as outlined in step 3 of our plan" assistant: "Let me use the…