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 OutlineDriven/outline-driven-development --skill baremetal-startupgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/baremetal-startup)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/baremetal-startup"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/baremetal-startup/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/outlinedriven/outline-driven-development/baremetal-startup"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/baremetal-startup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00057 | $0.02068 |
| Opus 5 | $0.00028 | $0.01034 |
| Sonnet 5 | $0.00011 | $0.00414 |
| Haiku 4.5 | $0.00006 | $0.00207 |
Grade A, and why
baremetal-startup 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 3d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bare-metal startup
Contract
| Field | Bound contract |
|---|---|
| Trigger | Firmware never reaches main(), a new MCU is being brought up without a vendor HAL, or a custom startup.s, Reset_Handler, or crt0 is being written or debugged on Cortex-M or RISC-V. |
| Authority | Read-only: emits startup code, linker symbol requirements, and diagnostics to chat; the user places them in the project. Rollback is not needed because no file is written. No remote mutation. |
| Side effect | Chat output only. |
| Done | The startup path from the reset vector to main() is stated step by step, the vector table layout and the .data and .bss loops are given for the target core, and every linker symbol the code uses is listed with the linker script line that defines it. |
Inputs
- Target core (Cortex-M0+, M3, M4F, M7, or RISC-V RV32 or RV64) and the vendor part.
- The linker script, or at least its
MEMORYregions and the symbols it exports. - Whether the firmware boots directly from flash or is launched by a bootloader at an offset.
- Which C runtime is linked (newlib, newlib-nano, none) and whether C++ static constructors exist.
Procedure
-
State the boot sequence for the core before writing code. On Cortex-M the hardware loads
SPfrom vector 0 and branches to vector 1,Reset_Handler. On RISC-V the core starts at the reset address withspundefined, so the first instruction sets it. Done when: the sequence is written as an ordered list from power-on tomain().Cortex-M vector table head:
Index Entry 0 Initial stack pointer ( _estack)1 Reset_Handler2 NMI_Handler3 HardFault_Handler4 and up Fault handlers, then device IRQ handlers in NVIC order -
Write the vector table and
Reset_Handlerfor Cortex-M. Copy.datafrom its load address in flash to its run address in RAM, zero.bss, then callSystemInitandmain. Compare addresses with an unsigned branch (bcc), because addresses are unsigned. Done when: the assembly assembles for the target and the three symbol pairs come from the linker script.
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.
- 3d ago First seen · 149 lines · 57 tokens per session scan A ed16c0b1c227
baremetal-startup is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed 4d ago), licensed Apache-2.0. It adds 57 tokens to every session and 2,068 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 2026-09-06.
Other skills, from other repositories
gke-ai-troubleshooting-tpu-dynamic-slices-monitoring
Monitors, troubleshoots, and manages GKE TPU Dynamic Slices custom resources. Use when checking TPU slice lifecycle states, troubleshooting slice provisioning failures, validating single-slice or multi-slice (JobSet) workload manifests, or safely patching stuck finalizers and disabling the slice controller. Don't use…
doca-flow
Build and debug DOCA Flow applications on supported NVIDIA NICs/DPUs: define match/action pipes, initialize ports and representors, choose forwarding targets, validate pipes before hardware programming, read counters, match the Flow version to the installed DOCA release, and diagnose Flow API errors. Trigger on DOCA…
evpn-vxlan-fabric
EVPN/VXLAN fabric audit and troubleshooting — VTEPs, VNIs, route types, multihoming, underlay/overlay validation. Use when troubleshooting VXLAN overlay reachability, auditing leaf-spine fabric health, debugging silent hosts or asymmetric flooding, or validating anycast gateway and ESI multihoming state.
triaging-robot-logs
Use when investigating a robot log or flight log with bagel — finding events, anomalies, or "what happened around X" in rosbags, MCAP, CAN, MDF4, ULog, or Betaflight files. Routes to the server's triage capability.
yam-robot-debugging
Use for YAM real-robot debugging: saved-script edits, physical runs, artifact inspection, autonomous patch-and-retry loops, and compact runtime context updates.
diagnose-hid-keycodes
Find out what an unknown USB/Bluetooth HID button actually emits without assuming the label or documentation. Uses Karabiner's.