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 jschulte/claude-plugins --skill arch-auditgit clone --depth 1 https://github.com/jschulte/claude-pluginsWrote 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/jschulte/claude-plugins/arch-audit)<a href="https://agentmods.dev/skills/jschulte/claude-plugins/arch-audit"><img src="https://agentmods.dev/badge/skills/jschulte/claude-plugins/arch-audit/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/jschulte/claude-plugins/arch-audit"><img src="https://agentmods.dev/badge/skills/jschulte/claude-plugins/arch-audit.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 | $0.00067 | $0.04061 |
| Opus 5.5 | $0.00027 | $0.01624 |
| Sonnet 5.5 | $0.00013 | $0.00812 |
| Haiku 4.5 | $0.00007 | $0.00406 |
Grade A, and why
arch-audit 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 24d 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 — 408 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architectural Invariant Audit
You are performing an exhaustive architectural invariant sweep. This is fundamentally different from a code review or health audit:
- A code review asks "does this look right?" (subjective, sample-based)
- A health audit asks "how healthy is this?" (evidence-backed, scored)
- An arch-audit asks "where are the established patterns violated?" (exhaustive, mechanical, objective)
Your job: discover what conventions this codebase has established, define what "correct" looks like for each one, then check every single instance — not a sample, not what catches your eye, but ALL of them.
Core Principle
"A codebase with 300 known bugs and one with 300 unknown bugs differ only in whether someone asked the right questions."
This skill IS the right questions, systematized.
How This Differs from /health-audit
| Dimension | health-audit | arch-audit |
|---|---|---|
| Mode | Read code, evaluate quality | Discover patterns, check every instance |
| Finds | Whatever catches the eye | Every violation of every invariant |
| Misses | Absences (missing filter, missing arg) | Instances its enumeration search cannot see (dynamic dispatch, generated code, aliases) |
| Output | Scored report with findings | Violation register with fix tracks |
| Best for | "How's the code?" | "What's silently broken?" |
Recommended workflow: Run health-audit first to diagnose systemic issues, then arch-audit to enumerate every instance. They're complementary.
Focus Prompt Support
This skill accepts an optional focus prompt:
- No focus: audit the entire codebase across all pattern classes
- Focused:
"auth patterns","soft-delete","error handling in webhooks","state machine coverage" - Stack-specific:
"Prisma queries","Next.js API routes","React hooks"
When focused, skip Phase 1 pattern classes outside the scope and go deep on the requested area.
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.
- 24d ago First seen · 408 lines · 67 tokens per session scan A 3b79850ca1af
arch-audit is a skill published in the GitHub repository jschulte/claude-plugins (0 stars, last pushed 25d ago), licensed MIT. It adds 67 tokens to every session and 4,061 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-12.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…