device-driver-development — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited device-driver-development (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
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.
references/knowledge-map.md for the driver domain map and conceptual model./dev, sysfs, debugfs, ioctl, mmap, netdev, IIO buffers, input events, DT binding, or kernel consumer API.devm_*, explicit unwind only when ordering or shared state requires it.tasks/bringup-workflow.md.tasks/implementation-patterns.md and references/subsystem-matrix.md.references/kernel-engineering-principles.md.tasks/debugging-testing.md.references/tooling-matrix.md.tasks/step-verify-workflow.md.tasks/upstream-maintenance.md.references/review-checklist.md.copy_{to,from}_user() only at ABI boundaries and handle partial copy/error paths.devm_ioremap_resource() plus accessor APIs; do not dereference __iomem.scripts/driver_static_scan.py <path> for heuristic review of C, Kconfig, Makefile, DTS, and binding files.scripts/module_smoke_build.sh <module-dir> [kernel-build-dir] for out-of-tree build smoke checks.scripts/tooling_probe.sh to inspect local build, static-analysis, tracing, fuzzing, QEMU, and KGDB tool availability.sparse, Smatch, Coccinelle, checkpatch.pl, dt_binding_check, and subsystem selftests.dmesg, /proc/interrupts, sysfs state, tracepoints/ftrace/bpftrace, bus transactions, and userspace API results.When answering, name assumptions, selected subsystem, key APIs, file layout, validation commands, and residual hardware risks. If source evidence is missing, say what must be measured or read next.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.