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 skills add thixpin/pitway --skill architecture-reviewgit clone --depth 1 https://github.com/thixpin/pitwayWrote 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/skills/thixpin/pitway/architecture-review)<a href="https://agentmods.dev/skills/thixpin/pitway/architecture-review"><img src="https://agentmods.dev/badge/skills/thixpin/pitway/architecture-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/thixpin/pitway/architecture-review"><img src="https://agentmods.dev/badge/skills/thixpin/pitway/architecture-review.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.00684 |
| Opus 5 | $0.00000 | $0.00342 |
| Sonnet 5 | $0.00000 | $0.00137 |
| Haiku 4.5 | $0.00000 | $0.00068 |
Grade A, and why
architecture-review 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 9d 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architecture Review
Judge structure by how safely the system can change, not by resemblance to a textbook pattern. Recommend the smallest set of moves that improves changeability; never propose a rewrite when an incremental path exists.
Scope
Reviews the system — modules, layers, and dependencies across the codebase.
Use for structural assessment: technical debt, module boundaries, coupling, duplication across modules, planning a refactor.
Do not use for:
- Reviewing a single diff or file for quality — use
code-quality-review. The same concerns (DRY, single responsibility) appear in both; the difference is altitude. Within one change, they are that skill's; across modules, they are this one's. - Implementing the refactor. This skill produces recommendations; carrying them out is ordinary work under
CLAUDE.md.
Method
- Map before judging. Identify the modules/layers, what depends on what, and where the domain logic actually lives. Use the dependency direction the code has, not the one the docs claim.
- Find where change hurts. Ask of recent or likely changes: how many places must be touched? Files that always change together but live far apart, or single files that change for unrelated reasons, mark the real problems.
- Diagnose against the checklist below.
- Recommend minimally. Each recommendation: the problem, the concrete evidence (files/modules), the smallest fix, and what it unblocks. Order by pain relieved per effort. Explicitly list what should be left alone.
Checklist
Duplication
- Same business rule or knowledge implemented in multiple places (the dangerous kind — they drift apart).
- Copy-paste variants of one workflow differing only in details that should be parameters.
- Do not flag incidental similarity between things that change for different reasons; merging those creates coupling, not reuse.
Single responsibility
- Modules or classes that change for unrelated reasons (e.g., a "service" doing validation, persistence, formatting, and notifications).
- God objects most of the codebase imports; files disproportionately large for their layer.
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.
- 9d ago First seen · 53 lines · 0 tokens per session scan A d3c79781c122
architecture-review is a skill published in the GitHub repository thixpin/pitway (19 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 684 tokens. 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 2026-08-30.
Other skills, from other repositories
audit
Use for PACEflow internal full audit: run five independent review agents, verify evidence from code/tests/logs, de-duplicate findings, and produce a severity-ranked report for release gates or comprehensive code review.
orchestrate
Coordinate an active Hardproof run across discovery, design, planning, implementation, review, verification, delivery, and learning.
review
Challenge a Hardproof implementation during REVIEW against its approved contract, code quality, tests, and risk boundaries.
behavior-implement
Implement behavior through a red-green-refactor cycle when focused automated tests are proportionate. Use for product behavior changes and for engineering tooling and infrastructure only when native checks are insufficient and concrete complexity or failure risk warrants dedicated tests.
link-check
Verify @file references in AIWG skills and agents against the linking contract — per-file or corpus-wide, with optional auto-fix.
agentplane-task-closure-recovery
Use when Agentplane task completion, direct finish, branchpr integration, hosted-close, close-tail PRs, PR metadata, dirty task artifacts, or remote branch divergence need diagnosis or recovery.