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 skills add wedsamuel1230/arduino-skills --skill memory-budgetinggit clone --depth 1 https://github.com/wedsamuel1230/arduino-skillsWrote 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/wedsamuel1230/arduino-skills/memory-budgeting)<a href="https://agentmods.dev/skills/wedsamuel1230/arduino-skills/memory-budgeting"><img src="https://agentmods.dev/badge/skills/wedsamuel1230/arduino-skills/memory-budgeting/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wedsamuel1230/arduino-skills/memory-budgeting"><img src="https://agentmods.dev/badge/skills/wedsamuel1230/arduino-skills/memory-budgeting.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00050 | $0.00624 |
| Opus 5 | $0.00025 | $0.00312 |
| Sonnet 5 | $0.00010 | $0.00125 |
| Haiku 4.5 | $0.00005 | $0.00062 |
Grade A, and why
memory-budgeting 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 12d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Budgeting
Make memory a measured budget, not a nominal board label. The implementation is embedded C/C++; scripts may parse build or runtime logs when that is the useful deterministic tool.
Intake
Record exact board/module, flash/RAM/EEPROM/PSRAM, bootloader and partition layout, framework, compiler flags, libraries, static buffers, task stacks, filesystem/OTA needs, peak protocol sizes, and available measurement output.
Process
- Separate static data, code/flash, heap, task stacks, interrupt state, filesystem, bootloader, and OTA/update slots.
- Establish a numeric budget and minimum margin for every category. Use board references rather than copying a family-level number into a project claim.
- Capture compiler size output and map changes to a feature or dependency.
- Measure runtime free heap, largest block, stack high-water marks, and reset/ allocation failures under representative load where the framework exposes them.
- On AVR, keep literals in flash with
F()where appropriate, prefer bounded buffers and fixed-size storage, and avoid unbounded dynamicStringuse. On ESP32/RP2040, still bound buffers and distinguish heap, stack, flash, and filesystem. - Re-run the stress case after each memory-affecting change and record the tradeoff.
Anti-rationalization
| Shortcut | Response |
|---|---|
| "The datasheet says 520 KB, so memory is fine." | Separate usable heap, stacks, reserved regions, and radio/RTOS use. |
| "It compiled." | Inspect size output and exercise runtime allocation paths. |
| "Make the buffer bigger." | Bound the protocol and prove the required maximum. |
| "Use String everywhere." | Check fragmentation and lifetime; use bounded alternatives where needed. |
| "OTA only needs the application image." | Budget both update slots, bootloader, metadata, and rollback image. |
Verification
- Build-size output and runtime memory evidence name the exact target/version.
- Each budget has a margin and a failure response.
- Stress tests cover peak message, sensor, logging, radio, and task load as applicable.
- Optimization does not remove bounds, safety checks, or recovery behavior.
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.
- 12d ago First seen · 59 lines · 50 tokens per session scan A 1b579b94ebdb
memory-budgeting is a skill published in the GitHub repository wedsamuel1230/arduino-skills (21 stars, last pushed 23d ago), licensed MIT. It adds 50 tokens to every session and 624 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 2026-08-30.
Other skills, from other repositories
gke-ai-troubleshooting-tpu-dynamic-slices-monitoring
Monitors, troubleshoots, and manages GKE TPU Dynamic Slices custom resources. Use when checking TPU slice lifecycle states, troubleshooting slice provisioning failures, validating single-slice or multi-slice (JobSet) workload manifests, or safely patching stuck finalizers and disabling the slice controller. Don't use…
gke-ai-troubleshooting-tpu-vbar-oom
Diagnoses and prevents vbarcontrolagent segfaults, out-of-memory (OOM) errors, and TPU device initialization failures on TPU v6e nodes in GKE caused by race conditions during TPU device resets or high-frequency metrics polling. Use when troubleshooting vbarcontrolagent crashes, memory cgroup OOMs in serial console…
analyze-kernel-bottleneck
Systematically identify whether a GPU kernel is compute-bound, memory-bound, or latency-bound using roofline analysis, occupancy calculations, compute/load ratio per tile, and SASS instruction inspection. Produces a decision matrix for optimization strategy selection (cp.async, warp interleaving, tiling…
mcubuddy
Use when debugging MCU firmware or boards with McuBuddy, including probe, boot, HardFault, peripheral, register, memory, ELF/SVD, RTOS, RTT/UART, Keil, GDB, pyOCD, J-Link, ST-Link, or CMSIS-DAP work.
cleverence-mslx
A guide for editing Cleverence Mobile SMARTS configuration files used by handheld warehouse and shop terminals. Each .mslx file describes a workflow as a sequence of actions connected like a directed graph.
fpga-hardware-design-and-review-guide
Comprehensive FPGA hardware design guide based on real-world project experience. Covers pipeline architecture, timing optimization, SystemVerilog coding patterns, CDC handling, and practical debugging techniques. Use this skill when: (1) designing FPGA modules with timing constraints, (2) implementing video processing…