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 agentmods add skills/oghie/skillsets/device-driver-developmentnpx skills add oghie/skillsets --skill device-driver-developmentgit clone --depth 1 https://github.com/oghie/skillsetsWrote 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/oghie/skillsets/device-driver-development)<a href="https://agentmods.dev/skills/oghie/skillsets/device-driver-development"><img src="https://agentmods.dev/badge/skills/oghie/skillsets/device-driver-development.svg" alt="Measured on agentmods" 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 | $0.00052 | $0.00986 |
| Opus 5 | $0.00026 | $0.00493 |
| Sonnet 5 | $0.00010 | $0.00197 |
| Haiku 4.5 | $0.00005 | $0.00099 |
Grade A, and why
device-driver-development 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 5d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Device Driver Development
Core Rule
Treat every driver task as hardware contract engineering: identify the kernel version, hardware bus, subsystem owner, execution context, lifetime model, and userspace ABI before writing code.
Knowledge Grounding
- Use
references/knowledge-map.mdfor the driver domain map and conceptual model. - Treat API details as version-sensitive; verify callback signatures, helpers, and bindings against the active kernel tree or current kernel docs.
First Pass
- Classify the device: char, platform, I2C, SPI, regmap, IIO, GPIO, IRQ controller, input, RTC, PWM, regulator, framebuffer, or NIC.
- Identify interface surfaces:
/dev, sysfs, debugfs, ioctl, mmap, netdev, IIO buffers, input events, DT binding, or kernel consumer API. - Choose probe/remove lifetime: prefer
devm_*, explicit unwind only when ordering or shared state requires it. - Decide concurrency boundaries: process context, atomic context, IRQ top half, threaded IRQ, workqueue, timer, DMA callback.
- Define evidence needed: datasheet registers, bus transaction examples, DT node, logs, scope/logic-analyzer traces, and test plan.
Required Reads By Task
- New driver or hardware bring-up:
tasks/bringup-workflow.md. - Implementing subsystem callbacks:
tasks/implementation-patterns.mdandreferences/subsystem-matrix.md. - Memory, MMIO, mmap, DMA, IRQ, locking:
references/kernel-engineering-principles.md. - Debugging, validation, or intermittent behavior:
tasks/debugging-testing.md. - Toolchain, QEMU, KGDB, bpftrace, Smatch, Coccinelle, Syzkaller, Bootlin:
references/tooling-matrix.md. - Auditable engineering flow:
tasks/step-verify-workflow.md. - Upstream-ready patches or long-term maintenance:
tasks/upstream-maintenance.md. - Final review before handoff:
references/review-checklist.md.
Design Heuristics
- Prefer an existing kernel subsystem over a private char driver when the device matches a framework.
- Keep policy in userspace and mechanism in kernel; avoid inventing ABI without necessity.
- Never sleep in atomic context; never hold spinlocks across bus calls that may sleep.
- Use
copy_{to,from}_user()only at ABI boundaries and handle partial copy/error paths. - For MMIO use
devm_ioremap_resource()plus accessor APIs; do not dereference__iomem. - For register devices on I2C/SPI, consider regmap before custom read/write helpers.
- For DT, model hardware, not Linux driver internals; document compatibles and supplies/clocks/IRQs/GPIOs.
What ships with it
15 files 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.
- agent.md 3.6 KB
- agents/openai.yaml 275 B
- references/kernel-engineering-principles.md 3.8 KB
- references/knowledge-map.md 2.9 KB
- references/review-checklist.md 2.7 KB
- references/subsystem-matrix.md 3.4 KB
- references/tooling-matrix.md 4.0 KB
- scripts/driver_static_scan.py 6.1 KB runs code
- scripts/module_smoke_build.sh 848 B runs code
- scripts/tooling_probe.sh 685 B runs code
- tasks/bringup-workflow.md 2.6 KB
- tasks/debugging-testing.md 3.2 KB
- tasks/implementation-patterns.md 2.8 KB
- tasks/step-verify-workflow.md 2.7 KB
- tasks/upstream-maintenance.md 1.8 KB
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.
- 5d ago First seen · 62 lines · 52 tokens per session scan A 26fa3f81eb95
device-driver-development is a skill published in the GitHub repository oghie/skillsets (9 stars, last pushed 1mo ago), licensed MIT. It adds 52 tokens to every session and 986 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 2026-08-31.
Other skills, from other repositories
person-watchdog
摄像头人形监控守护（PersonWatchdog）的部署、运维与排障技能。用于在 Windows 上用前置摄像头持续检测"有人经过电脑前"，通过 Hermes 的 hermes send 向飞书私聊发送文字提醒 + 清晰人脸抓拍 + 停留时长；支持开机自启（计划任务 + supervisor 守护）、省电优化（运动门控、ONNX 限线程）、自动学习强化（误报区域抑制、自适应置信度阈值），并沉淀相机监控类项目的整套实测调试方法论（进程消失、CPU 飙高、ONNX 线程、画面亮度诊断、计划任务 LastRun=1999 等）。触发场景：用户说"帮我装个摄像头监控""检测有人经过电脑前发飞书""开机自启不生效""watchdog…
gke-compute-classes
Configures, optimizes, and troubleshoots GKE ComputeClasses. Use when configuring Spot VMs with on-demand fallback, targeting specific accelerators (GPUs/TPUs) or machine families, restricting ComputeClass access, or debugging pending pods related to node pool auto-creation. Do not use for cluster-level Node Auto…
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.
offensive-z-wave
Z-Wave attack methodology — sniffing with Z-Force / EZ-Wave / RTL-SDR + ZniffMobile, S0 (legacy) network-key derivation flaw and key reuse, S2 (modern) ECDH commissioning analysis, replay/injection on unauthenticated nodes, default-key brute-force on test deployments, and home-automation hub pivots. Use when targeting…
hsb-flash
Flash the FPGA on an HSB board connected to an NVIDIA devkit. Supports HSB Lattice boards (FPGA versions 2407, 2412, 2507, 2510) and Leopard Imaging VB1940 "all-in-one" cameras (FPGA versions 2507, 2510). Uses release-specific YAML manifests and board-type-specific program commands. Lattice and VB1940 commands must…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.