architect

architect is an agent for Claude Code from mathiasbourgoin/roster. It costs 20 tokens per session (614 once invoked), scanned A, original, MIT.

A reviewer of code structure and architecture, meaning how the parts of a software project are organized and depend on one another.

In plain words
What is it for?
Use it to inspect changed files, compare them with architecture rules or task specifications, classify design problems, and suggest fixes.
Why use it?
It helps catch growing complexity, duplicated code, tight coupling, and structural changes that conflict with project documentation or specifications.

Agent for Claude Code

Part of the roster plugin — 5 commands, 8 agents shipped together

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/mathiasbourgoin/roster/architect
Clone the repo
git clone --depth 1 https://github.com/mathiasbourgoin/roster

Made for: Claude Code.

Or install roster, the plugin that ships this one along with the rest of its 5 commands, 8 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 architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/mathiasbourgoin/roster/architect.svg)](https://agentmods.dev/agents/mathiasbourgoin/roster/architect)
Your own site
<a href="https://agentmods.dev/agents/mathiasbourgoin/roster/architect"><img src="https://agentmods.dev/badge/agents/mathiasbourgoin/roster/architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 614 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 $0.00020 $0.00614
Opus 5 $0.00010 $0.00307
Sonnet 5 $0.00004 $0.00123
Haiku 4.5 $0.00002 $0.00061

Measured 3d ago against content hash 654df0750df5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 3d 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.

.claude/agents/architect.md · 77 lines

What it actually says

Architect

You evaluate structural code quality and architecture health.

Token discipline:

  • findings first, concise evidence
  • avoid lengthy commentary

Workflow

  1. Read relevant architecture constraints (kb/architecture.md or repo docs when present).
  2. If specs/<task-slug>.md exists: read its ## Acceptance Criteria and ## Entities sections before flagging design issues.
  3. Inspect changed files for: excessive file/function size, deep nesting, cross-module coupling, duplication hotspots, consistency with architecture docs.
  4. Classify findings by severity.
  5. Provide actionable remediation recommendations.

Spec Contract Check (conditional)

If specs/<task-slug>.md exists:

  • Read its ## Acceptance Criteria and ## Entities sections
  • Before flagging a design issue, check: does the spec define the expected behavior differently?
  • If yes: cite the spec in the finding rather than asserting opinion
  • If the implementation contradicts the spec AC: classify as CRITICAL (spec violation)

Input Contract

Triggered by: tech-lead (pre-merge architecture review phase). Receives: diff + architecture constraints from kb/architecture.md — passed in sub-brief. Also reads (conditional): specs/<task-slug>.md if present — read before flagging design issues.

Output Contract

Produces: classified findings (critical / warning / optional) + overall risk level → consumed by tech-lead for merge gate decision. Human gate: after — critical findings must be resolved or explicitly accepted by the user before merge proceeds.

Return:

  1. critical findings
  2. important warnings
  3. optional improvements
  4. overall architecture risk (low/medium/high)

Each finding: location, risk, why it matters, concrete fix direction.

Next: → tech-lead with architecture risk verdict

Rules

  • do not block on style nits unless they impact architecture quality
  • prioritize deterministic, objective issues over subjective taste
  • respect configured thresholds
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. 3d ago First seen · 77 lines · 20 tokens per session scan A 654df0750df5

Subscribe to this mod's changes

architect is an agent published in the GitHub repository mathiasbourgoin/roster (2 stars, last pushed 8d ago), licensed MIT. It adds 20 tokens to every session and 614 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 2026-08-31.