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 JAICHANGPARK/workshop-harness --skill cross-architecture-checkergit clone --depth 1 https://github.com/JAICHANGPARK/workshop-harnessWrote 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/jaichangpark/workshop-harness/cross-architecture-checker)<a href="https://agentmods.dev/skills/jaichangpark/workshop-harness/cross-architecture-checker"><img src="https://agentmods.dev/badge/skills/jaichangpark/workshop-harness/cross-architecture-checker/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/jaichangpark/workshop-harness/cross-architecture-checker"><img src="https://agentmods.dev/badge/skills/jaichangpark/workshop-harness/cross-architecture-checker.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 | $0.00076 | $0.01036 |
| Opus 5 | $0.00038 | $0.00518 |
| Sonnet 5 | $0.00015 | $0.00207 |
| Haiku 4.5 | $0.00008 | $0.00104 |
Grade B, and why
cross-architecture-checker scanned grade B with 1 finding 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 10d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| **Linux** (Debian/Ubuntu/Arch) | CUDA / ROCm / CPU | No GUI support on headless servers; permission issues with docker socket | Terminal-only commands (`ollama serve &`) and user group additions (`sudo usermod -aG dock How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cross-Architecture Checker Skill
Purpose
Workshop attendees bring a wide variety of hardware platforms (Apple Silicon M1-M4, Intel x86_64 Mac, Windows x86_64 / ARM64, Linux, ChromeOS). This skill performs automated and agentic audits of tech stacks against known architecture constraints, identifying compatibility risks before the workshop session and producing mandatory alternative fallback workflows.
Hardware & OS Compatibility Matrix
| Architecture / OS | Primary Acceleration | Known Risks & Constraints | Mandatory Fallback Strategy |
|---|---|---|---|
macOS Apple Silicon (arm64) |
Metal / Unified Memory | Minimal risks; high performance for local LLMs | Standard Ollama or MLX (mlx-lm) acceleration |
macOS Intel (x86_64) |
CPU-only (No Metal) | LM Studio crashes / freezes; high latency; no Metal support | Force Ollama CLI (ollama serve) with quantized 2B/4B models (gemma4:e2b) or cloud API |
| Windows 11 x86_64 | CUDA / DirectML / CPU | PowerShell script execution disabled by policy; WSL2 required for Docker | Provide PowerShell bypass scripts (Set-ExecutionPolicy -Scope Process Bypass) and native binary fallbacks |
| Windows ARM64 (Snapdragon) | NPU / DirectML | x64 emulation performance overhead; Docker driver conflicts | Use native ARM64 Ollama builds or cloud Gemini API |
| Linux (Debian/Ubuntu/Arch) | CUDA / ROCm / CPU | No GUI support on headless servers; permission issues with docker socket | Terminal-only commands (ollama serve &) and user group additions (sudo usermod -aG docker $USER) |
| ChromeOS (Linux Container) | Virtualized CPU | Low RAM allocation; no direct GPU pass-through | Use cloud API (Google AI Studio / GCP Gemini) or small 2B quantized models |
Tech Stack Audit Rules
When auditing a workshop tech stack string (e.g. lmstudio,docker,mlx,python), the agent must evaluate the following rules:
lmstudioin stack:- Risk: Intel Mac users will experience severe GUI lag or crashes due to lack of Metal support.
- Fix: Mandate Ollama CLI (
http://localhost:11434) as an alternative indocs/00-architecture-compatibility-matrix.md.
mlx/mlx-lmin stack:- Risk: Strictly restricted to Apple Silicon (
arm64). Fails instantly on Windows, Linux, and Intel Mac. - Fix: Provide PyTorch / HuggingFace transformers or Ollama fallbacks for non-Apple Silicon users.
- Risk: Strictly restricted to Apple Silicon (
dockerin stack:- Risk: Windows Home users without WSL2 or hypervisor enabled will fail to start Docker Desktop.
- Fix: Provide local native executable setup or cloud sandbox alternatives.
ollamain stack:- Risk: High RAM model selection (
12B+) crashes 8GB RAM laptops. - Fix: Provide model quantization tiers (
gemma4:e2bfor 8GB,gemma4:e4bfor 16GB,gemma4:e12bfor 32GB+).
- Risk: High RAM model selection (
gemini-live/audioin stack (Real-time Streaming & Fluttergemini_live):- Risk: Acoustic echo causes Voice Activity Detection (VAD) self-interruption loop; OS mic privacy permissions blocked.
- Fix: Mandate headphones/earphones in
RUNBOOK.mdand verify OS microphone permissions (macOS Privacy Settings, Windows App Permissions, FlutterRECORD_AUDIO).
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.
- 10d ago First seen · 63 lines · 76 tokens per session scan B cdfd5fae09f3
cross-architecture-checker is a skill published in the GitHub repository JAICHANGPARK/workshop-harness (5 stars, last pushed 9d ago), licensed MIT. It adds 76 tokens to every session and 1,036 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…