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/hermeticormus/libreembed-claude-code/memory-mgmt-patternsnpx skills add HermeticOrmus/LibreEmbed-Claude-Code --skill memory-mgmt-patternsgit 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 | $0.01354 |
| Opus 5 | $0.00000 | $0.00677 |
| Sonnet 5 | $0.00000 | $0.00271 |
| Haiku 4.5 | $0.00000 | $0.00135 |
Grade A, and why
memory-mgmt-patterns 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 — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.
memory-mgmt-patterns
Knowledge Base
Embedded memory management patterns for deterministic, fragmentation-free firmware.
Pattern 1: Static FreeRTOS Objects
Avoid dynamic allocation for RTOS objects. All objects pre-allocated at compile time.
/* Static task */
static StackType_t s_sensor_stack[256];
static StaticTask_t s_sensor_tcb;
static TaskHandle_t s_sensor_handle;
/* Static queue: 8 messages of 16 bytes each */
static uint8_t s_queue_storage[8 * 16];
static StaticQueue_t s_queue_struct;
static QueueHandle_t s_queue;
/* Static semaphore */
static StaticSemaphore_t s_sem_struct;
static SemaphoreHandle_t s_sem;
void rtos_objects_create(void)
{
s_sensor_handle = xTaskCreateStatic(
sensor_task, "sensor",
256U, NULL, TASK_PRIO_SENSOR,
s_sensor_stack, &s_sensor_tcb);
configASSERT(s_sensor_handle != NULL);
s_queue = xQueueCreateStatic(8, 16, s_queue_storage, &s_queue_struct);
configASSERT(s_queue != NULL);
s_sem = xSemaphoreCreateBinaryStatic(&s_sem_struct);
configASSERT(s_sem != NULL);
}
Required in FreeRTOSConfig.h: configSUPPORT_STATIC_ALLOCATION 1.
Pattern 2: Memory Pool for Variable-Frequency Messages
When messages arrive at variable rate, a pool prevents heap fragmentation:
/* Message pool: 32 messages, 128 bytes each */
typedef struct {
uint8_t data[120];
uint8_t len;
uint8_t type;
uint16_t seq;
} msg_t;
#define MSG_POOL_SIZE 32U
static msg_t s_msg_pool[MSG_POOL_SIZE];
static uint32_t s_pool_used_mask = 0U; /* Bitmask: bit N = msg N is in use */
msg_t *msg_alloc(void)
{
taskENTER_CRITICAL();
for (uint32_t i = 0; i < MSG_POOL_SIZE; i++) {
if (!(s_pool_used_mask & (1U << i))) {
s_pool_used_mask |= (1U << i);
taskEXIT_CRITICAL();
return &s_msg_pool[i];
}
}
taskEXIT_CRITICAL();
return NULL; /* Pool exhausted: caller must handle */
}
void msg_free(msg_t *m)
{
uint32_t idx = (uint32_t)(m - s_msg_pool);
configASSERT(idx < MSG_POOL_SIZE);
taskENTER_CRITICAL();
s_pool_used_mask &= ~(1U << idx);
taskEXIT_CRITICAL();
}
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 · 187 lines · 0 tokens per session scan A be141cf2e0b3
memory-mgmt-patterns is a skill 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 1,354 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 skills, from other repositories
hardware-iot-bus
Low-level I2C and SPI bus peripheral control for embedded Linux boards (Orange Pi, Raspberry Pi, RISC-V).
aether-iot-query
Use this skill when the user asks about a live AetherEdge runtime: channels, points, real-time values, history, alarms, rules, models, instances, routing, SHM health, service health, or system status. Use aether CLI commands to answer — do NOT inspect source code, local database files, or config YAMLs to answer…
aether-iot
Build, integrate, diagnose, or generate applications for the AetherEdge AI-native edge kernel. Use for AetherEdge onboarding, SDK compositions, device and topology clients, read-only operations UIs, MCP integration, Domain Packs, or governed IoT commands where live-state authority and physical-device safety must be…
avr-bare-metal-embedded
Curated Knowledge API for AI Agents — 68 MCP tools, 200+ skill packs, 46K chunks, semantic search over 670K vectors, 5-layer validation pipeline. Works with Claude Code, Cursor, Cline, Windsurf.
smart-product-dev
End-to-end IoT product development orchestration for TuyaOpen projects. Guides from requirements gathering → Tuya Platform product/DP creation → complete embedded firmware generation. State-machine: detects project state and picks up from wherever development currently stands.
edge-iot
Edge computing, IoT protocols, and embedded systems integration.