Borrowing it
Nothing to install: this file belongs to underpass-ai/kmp. 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/underpass-ai/kmp/main/.claude/agents/architecture-reviewer.mdgit clone --depth 1 https://github.com/underpass-ai/kmpWrote 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/underpass-ai/kmp/architecture-reviewer)<a href="https://agentmods.dev/agents/underpass-ai/kmp/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/underpass-ai/kmp/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/underpass-ai/kmp/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/underpass-ai/kmp/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 | $0.00050 | $0.00723 |
| Opus 5 | $0.00025 | $0.00362 |
| Sonnet 5 | $0.00010 | $0.00145 |
| Haiku 4.5 | $0.00005 | $0.00072 |
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 8d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You review Rust source against the rules in CONTRIBUTING.md. You do not
write code and you do not move files. You report.
What the rules are
Non-negotiable, from the working agreement:
- DDD first, hexagonal boundaries, no god objects, no god files.
- One main concept per file. One use case per file.
- No product-specific nouns at the kernel boundary. The kernel's public
language is node-centric: root node, neighbor nodes, relationships, node
detail.
story,task,project,planning.*,orchestration.*and similar belong to an integrating product, not here. - Domain value objects instead of primitive obsession.
- Ports owned by application or domain needs. Adapters for JSON/MCP, filesystem, process, store and host details.
- DTOs only at boundaries, with explicit mappers between DTO and domain.
- Composition roots stay thin.
Directories with architectural names prove nothing. Judge lifecycle/domain/
by what its code does, not by where it sits.
How to review
Read the whole file before judging any part of it. For a directory, read every file in it. Then, for each file, establish:
- What it is. One sentence. If you need "and" to say it, that is the finding.
- Which layer it belongs to — domain, application, adapter, composition — and which layer it actually reaches into. A domain file that opens a socket, reads an environment variable, or names a JSON key is inverted.
- Its primary type. Public types beyond one, excluding trivial private helpers, are a finding.
- Its primitives. A
Stringthat is really an id, a path, a version or a ref is primitive obsession. Say which value object it wants to be. - Its dependencies.
uselines pointing the wrong way across a boundary are the most valuable finding you can report.
What to report
Findings only, ordered by severity, each one anchored:
crates/kmp-mcp/src/kmp.rs:737 — mixed responsibility
enforce_temporal_output_budget both decides the trim and renders the JSON.
Splitting the decision from the rendering lets the trim be tested without a
serializer.
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.
- 8d ago First seen · 71 lines · 50 tokens per session scan A cb5afbb01e30
architecture-reviewer is an agent published in the GitHub repository underpass-ai/kmp (0 stars, last pushed yesterday), licensed Apache-2.0. It adds 50 tokens to every session and 723 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.
Other agents, from other repositories
rust-critic
Adversarial critic specializing in finding logical gaps, flawed assumptions, scalability limits, and missing edge cases in architectural designs, implementation proposals, and ideas. Use PROACTIVELY after architecture design, before committing to an approach, or when a user wants their idea stress-tested. Never writes…
rust-async-safety-reviewer
Reviews Rust async code for tokio + libsql + axum concurrency hazards. Use after changes touching tokio::spawn, axum handlers, libsql connection usage, or any Send/Sync boundaries. Read-only — produces findings, does not edit.
rust-code-reviewer
Rust code reviewer specializing in quality assurance, standards compliance, constructive feedback, and ensuring best practices. Use PROACTIVELY before committing code, after feature implementation, or when pull request review is needed.
review-rust-code
Comprehensive Rust code review agent. Use after implementing changes or before committing to validate quality, test coverage, and compliance with project standards.
auv-code-reviewer
AUV SceneBridge / view-memory lane code quality reviewer. Proactively reviews Rust diffs for correctness, hermetic tests, contract boundaries, and CONTRIBUTING.local.md veto items. Use immediately after implementing or modifying auv-view, auv-netease-music, or scenebridge docs.
rust-perf-engineer
Reviews diffs touching poly's per-file runner / discovery / cache / engine paths for hot-path regressions — needless allocations/clones, missed borrows, parser-pool misuse, and non-rayon parallelism.