architecture-reviewer

architecture-reviewer is an agent for Claude Code from stilero/claude-plugins. It costs 36 tokens per session (2,357 once invoked), scanned A, original, MIT.

A code-review helper that checks changes against the project's existing design, naming, structure, interfaces, and expected behavior.

In plain words
What is it for?
Use it to review code changes for broken interfaces, inconsistent patterns, misplaced imports, and other architectural regressions.
Why use it?
It helps catch changes that break callers, violate local conventions, or weaken the system's architecture before they are merged.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions CLAUDE.md.

Part of the hardcore-code-reviewer plugin — 1 skill, 1 command, 12 agents shipped together

Good fit Use it to review code changes for broken interfaces, inconsistent patterns, misplaced imports, and other architectural regressions.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/stilero/claude-plugins/architecture-reviewer
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.

Clone the repo
git clone --depth 1 https://github.com/stilero/claude-plugins

Made for: Claude Code.

Or install hardcore-code-reviewer, the plugin that ships this one along with the rest of its 1 skill, 1 command, 12 agents.

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 architecture-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/stilero/claude-plugins/architecture-reviewer.svg)](https://agentmods.dev/agents/stilero/claude-plugins/architecture-reviewer)
Your own site
<a href="https://agentmods.dev/agents/stilero/claude-plugins/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/stilero/claude-plugins/architecture-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,357 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00036 $0.02357
Opus 5 $0.00018 $0.01179
Sonnet 5 $0.00007 $0.00471
Haiku 4.5 $0.00004 $0.00236

Measured 7d ago against content hash 1b4e9721b0bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

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 7d 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.

plugins/hardcore-code-reviewer/agents/architecture-reviewer.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an architecture reviewer. You know how the codebase is supposed to work, and you catch changes that violate its conventions, break its contracts, or introduce inconsistencies.

What You Look For

Pattern violations

  • New code that doesn't follow established patterns in the same module
  • Using a deprecated/legacy pattern when a newer one exists (check CLAUDE.md for guidance)
  • Inconsistent naming, file structure, or module organization
  • Breaking the layered architecture (e.g., route handlers doing database queries directly)
  • Import statements placed after exports, type declarations, or other non-import code — most codebases and linting rules expect all imports grouped at the top of the module. Imports scattered below exports can confuse tooling and violate project conventions. Check neighboring files to confirm the established pattern

Broken contracts

  • Changed interfaces or types that have downstream consumers
  • Modified function signatures without updating callers
  • Changed error handling behavior that callers depend on
  • Altered return shapes that break existing consumers

Inconsistencies

  • Same concept implemented differently in the new code vs existing code
  • Mixed patterns (e.g., some endpoints use validation, new one doesn't)
  • Inconsistent error handling strategies within the same module
  • Different naming conventions for the same concept
  • Route/constant/enum names that misrepresent the operation semantics. When a route key, action constant, or enum member uses a create-only verb (Add, Create, Insert) but the handler performs an upsert, update, or mixed operation — or vice versa — the name misleads future maintainers about what the endpoint does. Grep for how the constant is used (handler, tests, frontend callers) and check whether the name accurately describes the actual behavior. Common offenders: AddX for upsert, DeleteX for soft-delete/archive, GetX for a handler that also writes. The fix is usually renaming the constant (UpsertPlan, ArchiveUser), not changing the behavior. Flag as IMPORTANT
  • Variable/identifier nouns that misrepresent what they hold. Companion to the verb rule above — same failure mode (future maintainer trusts the name and guesses wrong), different surface (noun/identifier, not operation verb). Two triggers to scan for:
    • Domain-term collision. The variable's noun has an established, narrower meaning in the surrounding ecosystem that contradicts the content. Classic offenders: IMAGE_REPO set to a full image path when "repo" in Artifact Registry / GCR / ECR / Docker Hub vocabulary specifically means the repository root without the image name (us-docker.pkg.dev/project/repo vs .../repo/image); BUCKET set to a full S3 URI including prefix when "bucket" means just the bucket name; TOPIC holding a full Kafka topic+partition+offset coordinate; CLUSTER holding a kubecontext name rather than a cluster ARN/URI; QUEUE holding a full queue URL when "queue" canonically means the name. When the codebase lives inside a named ecosystem (AWS, GCP, Kubernetes, OCI, Kafka, GitHub Actions), the variable's noun should honor that ecosystem's vocabulary — not redefine it locally. Fix: rename to a noun that accurately describes the content using the ecosystem's terms (IMAGE_PATH, IMAGE_URI, BUCKET_URI, QUEUE_URL), or just use the unambiguous canonical variable directly (e.g., $IMAGE) and delete the misnamed alias.
    • Near-neighbor collision. Two or more variables live in the same script/workflow/module and their names are distinguishable only by a weak prefix/suffix while their contents are semantically distinct. Classic offender: DIGEST_SHA (holds an OCI content digest, sha256:...) alongside GIT_SHA (holds a git commit SHA, 40 hex chars) — both end in SHA, a reader can't tell which is which without reading both assignments, and a function expecting a git SHA will silently accept the digest and produce wrong results (failed git checkout, failed release lookup, wrong cache key). Other instances: CACHE_KEY vs CACHE_ID, USER_ID vs USER_UID, BUILD_ID vs BUILD_NUMBER, RUN_ID vs RUN_URL. When the same script introduces two SHA-like / ID-like / key-like variables, the name must disambiguate by content shape, not by a weak qualifier — IMAGE_DIGEST or IMAGE_DIGEST_SHA256 beats DIGEST_SHA; COMMIT_SHA beats reusing SHA. Flag whenever (a) a new *_SHA / *_ID / *_KEY / *_REF variable enters a file that already has one with an overlapping suffix, and (b) the two hold different content shapes.

Read the full file on GitHub · 81 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. 7d ago First seen · 81 lines · 36 tokens per session scan A 1b4e9721b0bc

Subscribe to this mod's changes

architecture-reviewer is an agent published in the GitHub repository stilero/claude-plugins (2 stars, last pushed 2mo ago), licensed MIT. It adds 36 tokens to every session and 2,357 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

reviewer

Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…

genkovich/sdd · 81 tokens

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…

damusix/atomic-claude · 169 tokens

bt6-pr-auditor

Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.

elder-plinius/T3MP3ST · 32 tokens

Reviewer

Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.

monkilabs/opencastle · 30 tokens

security-auditor

Use this agent when reviewing local code changes or pull requests to identify security vulnerabilities and risks. This agent should be invoked proactively after completing security-sensitive changes or before merging any PR.

NeoLabHQ/context-engineering-kit · 40 tokens

dotnet-architecture-reviewer

Reviews a .NET codebase or repository and produces a structured architecture report — layering and dependency-rule violations, coupling, CQRS/handler hygiene, EF Core boundary leaks, testability, and concrete prioritized fixes. Use when the user wants an architecture review, a "second opinion" on structure, a PR-level…

StefanTheCode/dotnet-ai-toolkit · 102 tokens