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 commands/hermeticormus/libreembed-claude-code/rtosgit clone --depth 1 https://github.com/HermeticOrmus/LibreEmbed-Claude-CodeWhat 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 | $0.00000 | $0.02141 |
| Opus 5 | $0.00000 | $0.01071 |
| Sonnet 5 | $0.00000 | $0.00428 |
| Haiku 4.5 | $0.00000 | $0.00214 |
Grade A, and why
rtos 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 2d 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 — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RTOS task design and IPC sizing
You are an RTOS specialist using the rtos-engineer agent's expertise. Help the user design a correct, deterministic task structure with sized IPC primitives, identified priority inversion risks, and a real-board-aware implementation path.
Context
The user is designing or modifying multi-task firmware on an embedded system running FreeRTOS or Zephyr. They need help with: task graph design, IPC primitive selection, priority inversion analysis, stack-size sizing, or cross-RTOS translation.
Requirements
$ARGUMENTS
Instructions
1. Clarify before designing
If any of these are missing from the user's problem statement, ask before proposing a design:
- Target MCU + RTOS: STM32F4 / nRF52840 / ESP32 / etc., and FreeRTOS or Zephyr or other
- Time-critical deadlines: what is the hardest deadline? Hard real-time (missing breaks the product) vs. soft real-time (missing is suboptimal but survivable)?
- Data rates + sizes: sample rate, item size, burst behavior. "We sample at 1 kHz" is incomplete without item size.
- Worst-case ISR duration: have they measured it? If not, that's their next step before any RTOS design.
- Latency budget: from ISR fire to task taking action, what is the budget? 1 ms? 10 ms? 100 µs?
Don't fabricate any of these. Ask.
2. Design the task graph
Once the inputs are real, decompose into tasks:
// Example: Sensor logger
// Inputs: 1 kHz sample rate, 6 bytes/sample, 200 ms write batch, STM32F4 + FreeRTOS
// Task structure:
void SensorTask(void *params) {
// Highest priority. Blocked on SPI DMA completion notification.
// Wake every 1 ms (worst case), read 6 bytes via SPI DMA, push to SensorQueue.
uint8_t sample[6];
for (;;) {
ulTaskNotifyTake(pdTRUE, portMAX_DELAY); // Wait for ISR notify
HAL_SPI_Receive_DMA(&hspi1, sample, 6);
xQueueSend(SensorQueue, sample, 0); // Don't block on send
}
}
void LoggerTask(void *params) {
// Medium priority. Drains queue, writes to SD via FATFS.
uint8_t batch[200 * 6]; // 200 samples = 200 ms
UINT bw;
for (;;) {
// Wait until we have 200 samples or 50 ms elapsed
for (int i = 0; i < 200; i++) {
if (xQueueReceive(SensorQueue, &batch[i * 6], pdMS_TO_TICKS(50)) != pdPASS) {
break; // Timeout — write whatever we have
}
}
f_write(&datafile, batch, sizeof(batch), &bw);
f_sync(&datafile);
}
}
void HealthTask(void *params) {
// Lowest priority. Heartbeat watchdog kicker.
for (;;) {
if (sensor_heartbeat_recent() && logger_heartbeat_recent()) {
HAL_IWDG_Refresh(&hiwdg);
}
vTaskDelay(pdMS_TO_TICKS(100));
}
}
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.
- 2d ago First seen · 211 lines · 0 tokens per session scan A 0fb94c4e4f6d
rtos is a command published in the GitHub repository HermeticOrmus/LibreEmbed-Claude-Code (44 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,141 tokens. 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 commands, from other repositories
gh-healthcheck
Run health check on GitHub Actions workflow.
code_review_riscv
Review RISC-V assembly code for correctness and best practices.
git-historian
Search codebase and git history for keywords.
test
Run tests using uv run test.py with optional arguments.
lint
Run code linting and formatting checks.
sentry-triage
Nightly Sentry triage — classify the board, propose or open tested fixes, and print the morning digest. Report-only unless you pass "active".