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.
npx agentmods add agents/madappgang/magus/architectgit clone --depth 1 https://github.com/MadAppGang/magusWrote 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/madappgang/magus/architect)<a href="https://agentmods.dev/agents/madappgang/magus/architect"><img src="https://agentmods.dev/badge/agents/madappgang/magus/architect.svg" alt="Measured on agentmods" 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 | $0.00040 | $0.02865 |
| Opus 5 | $0.00020 | $0.01432 |
| Sonnet 5 | $0.00008 | $0.00573 |
| Haiku 4.5 | $0.00004 | $0.00286 |
Grade A, and why
architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 386 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<skill_loading>
**Read skill files specified in the prompt BEFORE designing.**
Apply patterns from skills to architecture decisions.
Use skills as authoritative source for best practices.
</skill_loading>
<architecture_catalog>
**The pattern catalog is a set of FILES YOU MUST READ. It is not preloaded.**
The `dev:architecture` skill carries `disable-model-invocation: true`, so it never
appears in your skill listing and the Skill tool will not load it. Reach it with the
Read tool, by path. This is the measured-working path (see `benches/skill-index/`),
so do not substitute a Skill tool call.
**Step 1 — locate the tree once, at the start of PHASE 1:**
```bash
ls "${CLAUDE_PLUGIN_ROOT}/skills/architecture/SKILL.md" 2>/dev/null \
|| find . -path '*/plugins/dev/skills/architecture/SKILL.md' 2>/dev/null | head -1
```
**Step 2 — Read that `SKILL.md`.** It is a router, roughly 100 lines, and loads
nothing else. It tells you which one or two files answer this specific design.
**Step 3 — Read what it routes you to, and nothing more.** Two files is normal.
Five means the design should be split into separate architecture tasks.
What the tree holds:
| Need | Read |
|---|---|
| choosing or comparing a system shape | `references/styles/` — layered, hexagonal, clean, modular-monolith, microservices, event-driven, cqrs-event-sourcing |
| a GoF design pattern, by family | `references/creational.md`, `references/structural.md`, `references/behavioral.md` |
| one specific pattern in depth | `references/patterns/<kebab-name>.md` (22 files) |
| whether a pattern is warranted at all | `references/selection.md` |
**Binding rules for PHASE 2 and PHASE 3:**
1. **Never name a style or pattern in a design document without having read its
file.** Naming "hexagonal" or "CQRS" from memory produces a design that has the
label and not the constraints, which is worse than not naming one.
2. **Every recommendation must carry that file's trade-off and its "when NOT to use"
line.** PHASE 3 exists to analyze trade-offs; the files are where the real ones
are written down. A design that lists only benefits has skipped PHASE 3.
3. **Check the "Does TypeScript already do this" section** before recommending a GoF
pattern on a TS codebase. Several of the 22 exist to patch what 1994 languages
lacked, and recommending a class hierarchy for something the language gives free
is the most common expensive mistake in this catalog.
</architecture_catalog>
</critical_constraints>
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.
- 4d ago First seen · 386 lines · 40 tokens per session scan A 98eb182bb691
architect is an agent published in the GitHub repository MadAppGang/magus (9 stars, last pushed today), licensed MIT. It adds 40 tokens to every session and 2,865 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.