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 HorizonRobotics/OE-Skills --skill j6-ucp-hbm-infergit clone --depth 1 https://github.com/HorizonRobotics/OE-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/horizonrobotics/oe-skills/j6-ucp-hbm-infer)<a href="https://agentmods.dev/skills/horizonrobotics/oe-skills/j6-ucp-hbm-infer"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-hbm-infer/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/horizonrobotics/oe-skills/j6-ucp-hbm-infer"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-hbm-infer.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 | $0.00139 | $0.03127 |
| Opus 5 | $0.00069 | $0.01563 |
| Sonnet 5 | $0.00028 | $0.00625 |
| Haiku 4.5 | $0.00014 | $0.00313 |
Grade A, and why
j6-ucp-hbm-infer 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 — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hbm_infer X86 Python Client Code Generation
Generate production-ready X86-side Python client code for hbm_infer — the Python SDK that connects to a BPU board via gRPC, deploys HBM models over SSH, runs inference, and manages board resources.
Quick Decision: Which Mode?
| Scenario | Mode | Module |
|---|---|---|
| Single model, single process | Standard | hbm_infer.hbm_rpc_session |
| Multi-process inference sharing one server | Flexible | hbm_infer.hbm_rpc_session_flexible |
| Multi-model pipeline with shared board resources | Flexible | hbm_infer.hbm_rpc_session_flexible |
Standard Mode — Quick Reference
from hbm_infer.hbm_rpc_session import HbmRpcSession
session = HbmRpcSession(
host="<board_ip>",
local_hbm_path="<hbm_file_path>",
)
# Use session, then always close
session.close_server()
HbmRpcSession.init parameters:
| Parameter | Type | Default | Description |
|---|---|---|---|
host |
str |
required | Board IP address |
local_hbm_path |
Union[str, List[str]] |
required | Local HBM file path(s) |
username |
str |
"root" |
Board SSH username |
password |
Optional[str] |
None |
Board SSH password |
ssh_port |
int |
22 |
SSH port |
remote_root |
str |
"/map/hbm_infer/" |
Board-side temp directory |
frame_timeout |
int |
90 |
gRPC per-frame timeout (seconds) |
server_timeout |
int |
5 |
Server auto-shutdown timeout (minutes) |
with_profile |
bool |
False |
Enable per-stage timing stats |
debug |
bool |
False |
Debug mode with extra logging |
compress_option |
str |
"NONE" |
gRPC compression: "NONE", "IN", "INOUT" |
core_id |
Union[int, List[int]] |
-1 |
BPU core ID(s), -1=CORE_ANY |
remote_environment |
Dict[str, Any] |
{} |
Board-side environment variables |
Key methods:
session(data, output_config=None, model_name=None)— Run inferencesession.get_model_names()→List[str]session.get_input_info(model_name=None)→Dict[str, Dict]session.get_output_info(model_name=None)→Dict[str, Dict]session.show_input_output_info(model_name=None)— Print model I/O infosession.get_profile(model_name=None)→ Cumulative timing stats (avg/min/max in ms)session.get_profile_last_frame(model_name=None)→ Last-frame timing stats (ms)session.close_server()— Must call explicitly to release board resources
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 229 lines · 139 tokens per session scan A b33d0e379f30
j6-ucp-hbm-infer is a skill published in the GitHub repository HorizonRobotics/OE-Skills (19 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 139 tokens to every session and 3,127 once invoked, about $0.0007 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
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-memory-thermal-ops
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
llama-cpp
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
minicpm5-deploy-vllm-ascend
Deploy MiniCPM5-2B with vLLM on Huawei Ascend NPU using vLLM-Ascend. Use when the user mentions vLLM-Ascend, Ascend NPU, Huawei Ascend, CANN, torchnpu, davinci devices, or wants an OpenAI-compatible MiniCPM5 server on Ascend hardware.
minicpm5-deploy-litert
Run MiniCPM5-2B or MiniCPM5-1B on-device with Google's LiteRT-LM runtime — the litert-lm CLI or its OpenAI-compatible server on a desktop, the Kotlin API or the AI Edge Gallery app on Android, the same .litertlm bundle on CPU or GPU. Use when the user says "LiteRT", "LiteRT-LM", "litertlm", ".litertlm", "Android"…