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 Midstall/claude-for-hardware --skill bare-metal-bringupgit clone --depth 1 https://github.com/Midstall/claude-for-hardwareWrote 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/midstall/claude-for-hardware/bare-metal-bringup)<a href="https://agentmods.dev/skills/midstall/claude-for-hardware/bare-metal-bringup"><img src="https://agentmods.dev/badge/skills/midstall/claude-for-hardware/bare-metal-bringup/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/midstall/claude-for-hardware/bare-metal-bringup"><img src="https://agentmods.dev/badge/skills/midstall/claude-for-hardware/bare-metal-bringup.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 | $0.00066 | $0.01078 |
| Opus 5 | $0.00033 | $0.00539 |
| Sonnet 5 | $0.00013 | $0.00216 |
| Haiku 4.5 | $0.00007 | $0.00108 |
Grade A, and why
bare-metal-bringup 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 11d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bare-Metal Bring-Up
Overview
Bringing up code on new silicon or a new architecture is a sequence of "does the most basic thing work yet" checkpoints. Each layer has a small number of mistakes that produce total silence or a single cryptic fault, and they are almost always init ordering, trap setup, address translation, or ABI mismatches.
Core principle: Get one character out the door first, then build up one checkpoint at a time. Until you have output, you are debugging blind, so the first job is always a working console, not the feature you wanted.
When to Use
- First boot on a new arch/board, or a port to a new target
- "Boots but hangs," silent boot, or a fault before main
- Traps/interrupts not firing, or firing into garbage
- Syscalls returning wrong values or corrupting registers across the boundary
- A driver works under one emulator/accelerator but not another (TCG vs KVM)
The Checkpoint Ladder
Climb in order. Don't debug a higher rung until the one below it is solid.
- Earliest output. Poke the UART directly (no driver, no allocator). One known byte. If you can't get a byte, nothing else is debuggable.
- Stack and BSS. A valid, correctly-placed stack and a zeroed BSS before any C/Zig/Rust runs. On some boots the stack must live inside a specific LOAD segment or the loader drops your initrd on top of it.
- Trap/exception vectors. Install the vector table, prove it by taking a deliberate trap and returning. Get this working before timers or interrupts.
- Timer. Architectural timer init MUST run before any kernel timer that divides by its frequency, or you divide by zero and panic with no output.
- Address translation. MMU/PMP/page tables. Identity-map what you need, then enable. Wrong here means a fault the instant translation turns on.
- Interrupts. Controller (PLIC/GIC/APIC), enable bits, the right per-IRQ vs global mask model for the part.
- Userspace / syscalls. Drop to user mode, take a syscall, return cleanly without clobbering caller-saved-by-the-ABI registers.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 58 lines · 66 tokens per session scan A c0e88d9f149d
bare-metal-bringup is a skill published in the GitHub repository Midstall/claude-for-hardware (20 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 66 tokens to every session and 1,078 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
tia-doctor
Manual, read-only prerequisite probe for TIA Portal V21 and its modular Openness API, with optional Python TIA Scripting and TIA MCP checks.
routeros-mac-telnet
MAC-Telnet protocol (MikroTik Layer-2 terminal/exec over UDP 20561) wire format, session handshake, and MD5 + MTWEI (EC-SRP) authentication. Use when: implementing or debugging a MAC-Telnet client/server, reaching a RouterOS device by MAC address without IP, parsing MAC-Telnet packets, understanding the WinBox-style…
routeros-syntax-inspection
Inspecting and validating RouterOS command/script syntax against a live device via /console/inspect (highlight, completion, syntax, child) and :parse IL. Use when: validating RouterOS commands before execution, explaining or linting RouterOS scripts, building syntax-aware tooling (LSP servers, validators, agent…
routeros-sniffer
RouterOS packet capture and TZSP streaming for protocol debugging. Use when: capturing packets on RouterOS, setting up /tool/sniffer, streaming live traffic via TZSP, using firewall mangle action=sniff-tzsp, debugging network protocols on MikroTik, receiving TZSP with Wireshark or tshark, saving pcap files from…
8051-mcu-debug
Use when debugging 8051-compatible microcontrollers, 51 MCU firmware, STC download issues, Keil C51 projects, interrupts, timers, UART, or startup failures.
cortex-m-debug
Use when debugging Cortex-M microcontrollers, firmware bring-up, SWD/JTAG sessions, faults, startup code, or flashing failures.