hailo-10h-llm-compiler: Skill for Claude Code

.claude/skills/device-debug/SKILL.md

device-debug is a skill for Claude Code from l-nmch/hailo-10h-llm-compiler. It costs 46 tokens per session (1,069 once invoked), scanned A, original, MIT.

A troubleshooting guide for Hailo-10 devices, which run their own firmware for tasks such as AI inference.

In plain words
What is it for?
Use it when Hailo-10 problems are reported only as generic failures and you need to inspect runtime, system-control, or neural-network logs.
Why use it?
It helps find the real cause behind vague client errors, timeouts, and hangs by checking the device's own logs first.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions CLAUDE.md.

This is l-nmch/hailo-10h-llm-compiler's own configuration. It tells Claude Code how to work on hailo-10h-llm-compiler itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything hailo-10h-llm-compiler configures →

Reuse

Borrowing it

Nothing to install: this file belongs to l-nmch/hailo-10h-llm-compiler. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/l-nmch/hailo-10h-llm-compiler/main/.claude/skills/device-debug/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/l-nmch/hailo-10h-llm-compiler

Made for: Claude Code.

Wrote 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.

agentmods badge for device-debug

README.md
[![agentmods](https://agentmods.dev/badge/skills/l-nmch/hailo-10h-llm-compiler/device-debug/github.svg)](https://agentmods.dev/skills/l-nmch/hailo-10h-llm-compiler/device-debug)
Your own site
<a href="https://agentmods.dev/skills/l-nmch/hailo-10h-llm-compiler/device-debug"><img src="https://agentmods.dev/badge/skills/l-nmch/hailo-10h-llm-compiler/device-debug/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.

agentmods 80×15 button for device-debug

Your own site · 80×15
<a href="https://agentmods.dev/skills/l-nmch/hailo-10h-llm-compiler/device-debug"><img src="https://agentmods.dev/badge/skills/l-nmch/hailo-10h-llm-compiler/device-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,069 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00046 $0.01069
Opus 5 $0.00023 $0.00535
Sonnet 5 $0.00009 $0.00214
Haiku 4.5 $0.00005 $0.00107

Measured 6d ago against content hash ed29ea80c96c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

device-debug 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 6d 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.

.claude/skills/device-debug/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Device-side debugging playbook

The most common trap in this project: a generic client-side error (HAILO_TIMEOUT, a hang, a vague failure code) hides a specific, immediately-legible error that only exists in the device's own logs. Chasing client-visible symptoms (file size, PCIe throughput, retry logic) before checking device-side logs has cost real time more than once — check this first, not last.

Why the client alone can lie to you

The Hailo-10H runs its own firmware with an embedded mini-Linux (visible in dmesg as u-boot-spl.bin / fitImage / image-fs loaded by the PCIe driver at boot). The "server" side of any client/server exchange (hailort.log talks about "sending N chunks to server") is that firmware, not a host process. It can fail immediately and specifically, then have its own generic timeout/error path fire seconds later — which is the only thing the client ever reports. If you only read the client log, you are debugging the echo, not the cause.

The technique

hailortcli logs runtime          # LLM/inference server logs — start here
hailortcli logs system_control    # control-plane / configuration logs
hailortcli logs nnc                # neural network core logs

Run this immediately after a failure, before forming hypotheses from client-side symptoms alone. Look for the first CHECK/error line, not the last — the device's own error-reporting path can itself time out and add a second, misleading failure a few seconds later.

General elimination order for opaque failures

  1. Device-side logs first (above) — often gives the exact failing check and file:line in one shot.
  2. Structural comparison against a known-good official artifact (hef_audit.py, parse-hef) — rules in/out the artifact itself before blaming the runtime.
  3. Client-side logs (HAILORT_CONSOLE_LOGGER_LEVEL=debug, hailort.log) — timing and chunking, useful once you know what to look for.
  4. Version alignment — driver, HailoRT, and server packages must come from the same software-suite drop; mixed versions fail in confusing ways. Check all three before deep-diving logic.

Read the full file on GitHub · 86 lines

Changes

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.

  1. 6d ago Changed · +13 lines ed29ea80c96c
  2. 10d ago First seen · 73 lines · 46 tokens per session scan A 95c48dafa2b4

Subscribe to this mod's changes

device-debug is a skill published in the GitHub repository l-nmch/hailo-10h-llm-compiler (14 stars, last pushed 9d ago), licensed MIT. It adds 46 tokens to every session and 1,069 once invoked, about $0.0002 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 2026-08-31.

Related

Other skills, from other repositories

trulens-instrumentation

Instrument LLM apps with TruLens OTEL-based tracing - from setup to debugging and optimization.

truera/trulens · 25 tokens

trulens-diagnosis

Diagnose low evaluation scores and generate actionable improvement recommendations.

truera/trulens · 17 tokens

bug-reproducer

Find likely software bugs in a codebase, rank concrete bug candidates, and prove or reject them with focused regression tests before proposing a fix. Also turn bug reports, stack traces, screenshots, failing behavior, support tickets, and regressions into minimal reproducible cases with red-to-green evidence. Use when…

AnastasiyaW/codex-claude-code-config · 147 tokens

refactoring-safely

Change the structure of code that already exists without changing what it does: smells as triggers, the named transformations (extract/inline, move feature, organise data, simplify conditionals), and above all the workflow that makes it safe — characterization tests first, one transformation at a time, green between…

AnastasiyaW/codex-claude-code-config · 221 tokens

code-complexity

Keep each unit comprehensible while the code grows: deep modules over shallow ones, information hiding, honest names, small functions with one job, error handling that does not lose the error, DRY and orthogonality, design by contract, and not leaving broken windows. Merges module-complexity management with…

AnastasiyaW/codex-claude-code-config · 252 tokens

kicad-review

Design review and validation workflow for KiCAD projects via MCP tools. Triggers on: "review my design", "check for errors", "audit", "DRC", "ERC", "find problems", "design review", "is this ready", "validate", "check my schematic", "check my PCB", "what's wrong", "run checks", "pre-fab review".

mixelpixx/Konnect · 83 tokens