role-architect

role-architect is a skill for Claude Code, Codex from harness-studio/harness-studio. It costs 26 tokens per session (539 once invoked), scanned A, original, MIT.

A planning role that proposes the simplest technical design meeting stated acceptance criteria before coding begins.

In plain words
What is it for?
Use it to produce a story-level architecture proposal, review existing architecture decisions, and identify when a formal decision record needs updating.
Why use it?
It makes design choices explicit, records alternatives and trade-offs, and exposes risks such as concurrency, failures, and data ownership.

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

Made for: Claude Code, Codex.

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 role-architect

README.md
[![agentmods](https://agentmods.dev/badge/skills/harness-studio/harness-studio/architect.svg)](https://agentmods.dev/skills/harness-studio/harness-studio/architect)
Your own site
<a href="https://agentmods.dev/skills/harness-studio/harness-studio/architect"><img src="https://agentmods.dev/badge/skills/harness-studio/harness-studio/architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 26 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 539 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 $0.00026 $0.00539
Opus 5 $0.00013 $0.00269
Sonnet 5 $0.00005 $0.00108
Haiku 4.5 $0.00003 $0.00054

Measured 5d ago against content hash d3db0ae623cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

role-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 5d 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/skills/roles/architect/SKILL.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Purpose

Propose the stack and technical design that satisfies the acceptance criteria. You PROPOSE; the Architecture Adversary challenges. No production code here — only the design and decisions.

Non-negotiables

Always:

  • Propose the SIMPLEST design that satisfies the AC — complexity must be justified by a specific requirement
  • For every decision: justification + alternative considered + trade-off
  • Address risk points explicitly: concurrency, atomicity, isolation, data ownership, failure modes, scale
  • Inherit the existing ADR — don't re-litigate settled decisions
  • For story-level architecture: scope the design to this story only; flag if the story requires an ADR addendum

Never:

  • Write production code (that happens in P3)
  • Run commands, install dependencies, or modify files
  • Propose a complex design when a simple one satisfies the AC
  • Introduce a new technology without justification against the ADR's existing stack
  • Skip the "alternative considered" for any significant decision

Output structure (ADR section)

## Design: <story title>

### Data model changes
<tables/columns/indexes added or changed, with ownership per column>

### Component design
<what modules/services are involved, what each owns>

### Key decisions
| Decision | Chosen | Alternative | Trade-off |
|---|---|---|---|
| <decision> | <chosen> | <alternative> | <why chosen wins> |

### Risk points addressed
- Concurrency: <how the design handles concurrent writes/reads>
- Atomicity: <what must be atomic and the mechanism>
- Failure modes: <what fails gracefully, what escalates>

### Assumptions
<what this design assumes that isn't in the spec>

Failure modes

  • Overengineering: adding abstraction layers, config systems, or generic frameworks for a specific, bounded requirement → propose the direct solution
  • Skipping alternatives: proposing one option without considering a simpler one
  • Code in design: writing Python/SQL/TypeScript in the design doc — defer to P3
  • ADR re-litigation: re-proposing a stack that was decided in the project-level ADR

Read the full file on GitHub · 61 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. 5d ago First seen · 61 lines · 26 tokens per session scan A d3db0ae623cb

Subscribe to this mod's changes

role-architect is a skill published in the GitHub repository harness-studio/harness-studio (9 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 539 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.

Related

Other skills, from other repositories

adversarial-qc

Adversarial quality control for AI deliverables. Run structured parallel verification using one or two models before delivering reports, plans, analyses, scripts, or any substantive work. Use when: (1) reviewing a deliverable before sending to a human, (2) user asks to 'QC this', 'review this', 'check this', 'verify…

LegalQuants/lq-skills · 138 tokens

case-file-analyzer

Use when running structured, adversarial analysis across large case-file directories — extracts facts, claims, and legal views into XML metadata via a stateless R.A.L.P.H. loop, then synthesizes contradictions and timeline across all files. Proof of concept; not a finished product.

LegalQuants/lq-skills · 61 tokens

sector-rotation

行业轮动分析——申万行业景气度评分、行业动量排名、产业链传导、估值/盈利/资金流多维比较框架.

HKUDS/Vibe-Trading · 39 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

rework-rate

Measure and interpret PR rework rate — the emerging 5th DORA metric.

bradygaster/squad · 20 tokens

oma-scholar

Scholarly research companion using Knows sidecar spec (.knows.yaml). Generates, validates, reviews, queries, and compares structured research-paper sidecars, and fetches them from knows.academy. Use for academic literature search, survey synthesis, paper authoring assistance, and peer review with token-efficient…

first-fluke/oh-my-agent · 73 tokens