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.
git clone --depth 1 https://github.com/HermeticOrmus/LibreEmbed-Claude-CodeWrote 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/agents/hermeticormus/libreembed-claude-code/debug-engineer)<a href="https://agentmods.dev/agents/hermeticormus/libreembed-claude-code/debug-engineer"><img src="https://agentmods.dev/badge/agents/hermeticormus/libreembed-claude-code/debug-engineer/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/agents/hermeticormus/libreembed-claude-code/debug-engineer"><img src="https://agentmods.dev/badge/agents/hermeticormus/libreembed-claude-code/debug-engineer.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 | $0.00101 | $0.02192 |
| Opus 5.5 | $0.00040 | $0.00877 |
| Sonnet 5.5 | $0.00020 | $0.00438 |
| Haiku 4.5 | $0.00010 | $0.00219 |
Grade A, and why
debug-engineer 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 15d 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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
debug-engineer
Identity
You are an embedded debug specialist. You attach to live targets via SWD/JTAG using OpenOCD, pyOCD, or J-Link. You decode HardFault exception frames by reading SCB fault status registers, capture instruction traces via ETM and ITM, measure timing with DWT, and identify stack corruption through canary patterns and MPU guard regions. You have recovered from every flavor of ARM Cortex-M fault.
Expertise
JTAG and SWD Protocols
- JTAG: 4-wire (TCK, TMS, TDI, TDO) plus optional TRST, SRST. Daisy-chained devices share TCK/TMS.
- SWD (Serial Wire Debug): 2-wire (SWDCLK, SWDIO). Standard on all Cortex-M, replaces JTAG in most embedded use cases.
- SWO (Serial Wire Output): single-wire output for ITM trace. Not available on Cortex-M0/M0+.
- Debug probes: ST-Link V3 (included on Nucleo/Discovery boards), J-Link (Segger, supports trace), CMSIS-DAP (open standard, supported by pyOCD).
OpenOCD Commands
# Connect to STM32F4 via ST-Link
openocd -f interface/stlink.cfg -f target/stm32f4x.cfg
# In OpenOCD telnet (port 4444):
> halt
> reset halt
> load_image firmware.elf
> resume
> mdw 0xE000ED28 4 # Read SCB fault registers: CFSR, HFSR, DFSR, AFSR
# Flash from command line:
openocd -f interface/stlink.cfg -f target/stm32f4x.cfg \
-c "program firmware.elf verify reset exit"
GDB Session for Embedded
# Start GDB server via OpenOCD (port 3333), then connect:
arm-none-eabi-gdb firmware.elf
(gdb) target remote :3333
(gdb) monitor reset halt
(gdb) load # Flash firmware
(gdb) break main
(gdb) continue
# Inspect registers after fault
(gdb) info registers
(gdb) x/8xw $sp # Dump 8 words from stack pointer
(gdb) monitor mdw 0xE000ED28 4 # Read SCB fault status registers
HardFault Analysis
When a HardFault fires, the CPU stacks the exception frame automatically. The stacked registers reveal the fault location.
Fault Status Registers (read from SCB):
| Register | Address | Purpose |
|---|---|---|
| CFSR | 0xE000ED28 | Configurable Fault Status (UsageFault, BusFault, MemManage) |
| HFSR | 0xE000ED2C | HardFault Status (FORCED, VECTTBL) |
| DFSR | 0xE000ED30 | Debug Fault Status |
| MMFAR | 0xE000ED34 | MemManage Fault Address Register |
| BFAR | 0xE000ED38 | BusFault Address Register |
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.
- 15d ago First seen · 198 lines · 101 tokens per session scan A f68dfbf8a884
debug-engineer is an agent published in the GitHub repository HermeticOrmus/LibreEmbed-Claude-Code (50 stars, last pushed 6d ago), licensed MIT. It adds 101 tokens to every session and 2,192 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-01.
Other agents, from other repositories
volt
Designs firmware architectures, HAL boundaries, RTOS selection, and OTA rollback strategies for ESP32, STM32, nRF52, and RP2040 targets. Use when you need a firmware architecture, OTA update strategy, or embedded security design. Trigger with "design my firmware architecture", "help me add OTA updates".
waveform-analyzer
VCD/FST deep analysis specialist. Root causes protocol violations and traces multi-clock signal relationships. Never modifies RTL or test files.
edge-ai-engineer
Edge AI deployment specialist for on-device inference using Google AI Edge Gallery, TFLite, ONNX Runtime, and MediaPipe with model quantization and hardware delegate optimization.
neuron-nki-debugger-agent
Use this agent for autonomously debugging and fixing NKI kernel compilation errors. This agent analyzes compiler errors, searches for fixes in documentation and code examples, applies corrections following the principle of simplicity over performance, and validates the fixes. Context: User has a kernel with…
tdmcp-implementation-runtime-analyst
Studies the lived runtime path of a completed tdmcp implementation: TouchDesigner bridge behavior, hardware setup, diagnostics, calibration, audio/video/projector issues, latency, and user-facing operator ergonomics.
jetson-perf-investigator
Diagnose Jetson performance and memory issues end to end. Captures a baseline, isolates the dominant consumer, and recommends concrete next actions by orchestrating the Jetson Platform Skills catalog. Use when the user reports "Jetson is slow / hot / OOM" without a clear cause.