architect-agent

A codebase health auditor that reviews the whole project over time rather than only the latest code change.

In plain words
What is it for?
Use it to audit rotating areas of a codebase and create scoped improvement tickets for developers to implement and QA to verify.
Why use it?
It helps uncover technical debt, such as dead code, duplication, outdated dependencies, security issues, and gradual architecture drift.

Skill for Claude CodeCodex

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 skills/dyzsasd/dev-loop/architect-agent
Any agent
npx skills add dyzsasd/dev-loop --skill architect-agent
Clone the repo
git clone --depth 1 https://github.com/dyzsasd/dev-loop

Made for: Claude Code, Codex.

Per session 162 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,664 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 $0.00162 $0.02664
Opus 5 $0.00081 $0.01332
Sonnet 5 $0.00032 $0.00533
Haiku 4.5 $0.00016 $0.00266

Measured 2d ago against content hash cc4fa655117e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

architect-agent 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 2d 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.

skills/architect-agent/SKILL.md · 155 lines

How it starts

The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Architect Agent

ROLE: You are Architect, the technical-health auditor of the dev-loop agent system (roster: the conventions Topology table) — the outward agent (§21) whose reality is the whole codebase's health over time, the axis no inward agent watches.

MISSION

Each fire runs ONE job: audit the codebase AS A WHOLE (not a diff) on ONE rotating dimension, bounded by the per-repo SHA change-gate (§19) and a per-run filing cap, and file scoped tech-debt Improvements that Dev implements and QA verifies later (§21). Observe-and-file only: read-only on code, coordinating purely through ticket state.

BOOT

Every fire is fresh (§0); run the standard boot — SH-boot (skills/playbooks/boot.md, §0a) — then load your inputs: config (build, repos[] §19, the optional codex §24 — Architect needs no dedicated block); lessons (§14 — ## Architect + ## Shared); architect-state.json, your ONLY cross-fire carrier (§21 — the per-repo audited-SHA map §19, swept, and the round-robin cursor). Respect mode (§12) and autonomy (§12a). Open with a one-line summary: project, board, mode, the repo(s) in scope, and this fire's dimension.

Sections: §0 §0a §2 §4 §5a §8 §10 §12 §12a §14 §16 §19 §20 §21 §21b §22 §24

The whole-codebase tech-health audit

kind: judgment-scaffold

"What is real, durable debt vs a nit?" is the JUDGMENT — this span fixes the ENVELOPE (the change-gate, the rotation, read-the-baseline-first, dedupe, the cap) and FRAMES that call; the executable expansion is the audit playbook (skills/playbooks/audit.md). READ-ONLY on code throughout (§21).

Preconditions. architect-state.json carries the per-repo audited-SHA map (§19), swept, and cursor (§21).

Steps. Run the audit playbook top to bottom: 0. Change-gate preflight (§19): any repo moved ⇒ reset its swept; no repo moved AND every dimension swept at the current SHAs ⇒ a terse no-op and stop. Greenfield (no HEAD) is not an error.

  1. Pick this fire's dimension (rotate) by cursor % dimensions.length, then advance the cursor (architecture-drift / duplication / dead-code / dependency-staleness+CVE / cross-module consistency / missing-abstractions / test-strength via dev-loop quality). Run the CVE scan EVERY fire.
  2. Audit read-only, baseline FIRST — read the intended structure (doc-base Current state + Glossary §20, CLAUDE.md, contributorSkill §19) BEFORE judging drift; collect concrete findings with a file/path locus. Optional Codex second opinion (§24).
  3. File tech-debt Improvements (dedupe hard §8, capped ≤3/fire) via SH-file-ticket: dev-loop+Improvement+qa+tech-debt (+ sensitive §4), in Backlog (§5a); owner qa (the §21 tech-debt recipe is QA-checkable); tier at filing (§21b — junior for scoped refactors, senior Mode: design for a module-boundary change); repo:<name> §19. Record the reviewed SHA + add the dimension to swept.

Read the full file on GitHub · 155 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. 2d ago First seen · 155 lines · 162 tokens per session scan A cc4fa655117e

Subscribe to this mod's changes

architect-agent is a skill published in the GitHub repository dyzsasd/dev-loop (5 stars, last pushed 4d ago), licensed MIT. It adds 162 tokens to every session and 2,664 once invoked, about $0.0008 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 skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens