RTOS kernel code review specialist. Use when user asks to review kernel code, audit driver implementation, check concurrency safety, review patch, 审查内核代码, 检查并发安全, review driver, audit ISR safety, check synchronization, or any request to review embedded/RTOS C code for correctness and safety.
How this skill is triggered — by the user, by Claude, or both
Slash command
/rtos-engineer-toolkit:kernel-reviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Perform systematic review of RTOS kernel / driver code with focus on correctness and safety.
Perform systematic review of RTOS kernel / driver code with focus on correctness and safety.
Read .claude/rtos-knowledge/coding-style.md for project-specific style rules. If not found, invoke project-init first.
For each file in the diff, check ALL of the following:
| Scenario | Correct approach |
|---|---|
| ISR ↔ task shared data | Disable interrupts or spinlock with IRQ save |
| Task ↔ task (short hold, no blocking) | Spinlock (SMP) or scheduler lock (UP) |
| Task ↔ task (may block) | Mutex or semaphore |
| Priority-sensitive | Priority-inheritance mutex |
| One-shot signal | Binary semaphore or event flags |
| Producer-consumer | Message queue or ring buffer |
Red flags:
Read rules from .claude/rtos-knowledge/coding-style.md and check compliance. If no knowledge file exists, infer from surrounding code in the same file.
## <filename>
### Line <N> [Block/Major/Minor/Nit]
<description of issue>
**Suggestion:** <fix or improvement>
After per-file review, provide:
npx claudepluginhub kaben123/rtos-engineer-toolkit --plugin rtos-engineer-toolkitGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.