Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/HorizonRobotics/OE-Skillsnpx agentmods add skills/horizonrobotics/oe-skills/j6-ucp-model-perf-evalWrote 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-model-perf-eval)<a href="https://agentmods.dev/skills/horizonrobotics/oe-skills/j6-ucp-model-perf-eval"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-model-perf-eval/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-model-perf-eval"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-model-perf-eval.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 | $0.00070 | $0.03556 |
| Opus 5 | $0.00035 | $0.01778 |
| Sonnet 5 | $0.00014 | $0.00711 |
| Haiku 4.5 | $0.00007 | $0.00356 |
Grade A, and why
j6-ucp-model-perf-eval 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 11d 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 — 286 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Perf Eval
通过 SSH/SCP 将 hrt_model_exec 部署到远程 J6 aarch64 开发板,遍历用户指定的 thread_num 和 core_id 参数组合运行 perf 评测,收集 --profile_path 结构化输出与 BPU/CPU 段级性能数据,生成性能汇总报告与最优配置推荐,并将报告保存到本地文件。
适用场景
- 在 J6 实板上评测 .hbm 模型推理性能(延迟、FPS)
- 扫描不同 thread_num / core_id 组合,探索最佳性能配置
- 远程部署 hrt_model_exec 到 aarch64 板端并执行 benchmark
- 对比不同配置的性能差异,生成结构化报告
- 分析模型 BPU/CPU 段的执行耗时占比
触发关键词:性能测试、perf、benchmark、性能评估、板端测试、模型吞吐量、参数扫描
工作流程
严格按照以下 6 步顺序执行,不可跳步。
Step 1:收集信息
向用户收集以下信息。缺少必填项时,必须逐一询问,全部收集完毕后方可进入 Step 2。
| 项目 | 必填 | 默认值 | 示例 |
|---|---|---|---|
| 板端 IP 地址 | 是 | - | 192.168.1.100 |
| SSH 用户名 | 是 | root |
root |
| SSH 端口 | 否 | 22 | |
| 认证方式 | 是 | - | 密码 或 密钥路径(如 ~/.ssh/id_rsa) |
| 模型文件路径 | 是 | - | /data/models/resnet50.hbm(板端)或本地路径 |
| model_name | 否 | 自动检测 | resnet50 |
| thread_num 范围 | 是 | 1,2,4,8 |
1-8 或 1,2,4,8,16 |
| core_id 选项 | 是 | 0 |
0,1 |
| frame_count | 否 | 200 | 500 |
| 板端部署目录 | 是 | /tmp/hrt_model_exec |
|
| enable_warmup | 否 | true | |
| enable_mem_lru | 否 | true |
必须明确确认的信息:
- 板端 IP 地址 — 若用户未提供,直接询问:"请提供开发板的 IP 地址"
- SSH 用户名 — 若用户未提供,询问:"SSH 登录用户名是什么?(默认 root)"
- 认证方式 — 若用户未提供,询问:"SSH 认证方式?密码或密钥路径"
- 模型文件路径 — 若用户未提供,询问:"请提供 .hbm 模型文件路径(板端路径或本地路径)"
- 模型文件位置 — 若用户只给了路径但未说明位置,询问:"该模型文件在本地还是已在板端?"
- 板端部署目录 — 若用户未指定,使用默认值
/tmp/hrt_model_exec,但需确认
thread_num 范围解析规则:
- 范围写法
1-8:展开为 1,2,3,4,5,6,7,8 - 列表写法
1,2,4,8:直接使用 - 展开后所有值必须在 [1, 32] 范围内
core_id 说明:
0= 任意核心(运行时自动选择)1= BPU Core 02= BPU Core 11,2= 双核并行- J6 有 2 个 BPU 核心,有效值:0, 1, 2, "1,2"
模型文件位置判断:
- 用户说明在板端:直接使用板端路径
- 用户提供本地路径:通过 SCP 上传到板端部署目录
Step 2:检查/构建 hrt_model_exec
- 检查本地
tools/hrt_model_exec/output_shared_J6_aarch64/目录是否存在且完整:- 必须存在:
aarch64/bin/hrt_model_exec - 必须存在:
aarch64/lib/下的 .so 文件
- 必须存在:
- 若目录不存在或不完整,执行构建:
cd tools/hrt_model_exec && bash build_aarch64.sh - 构建依赖
LINARO_GCC_ROOT环境变量指向交叉编译工具链。若构建失败,提示用户检查工具链配置。
Step 3:部署到远程板端
- SCP 部署 hrt_model_exec 工具目录:
scp -P <port> -r tools/hrt_model_exec/output_shared_J6_aarch64/ <user>@<ip>:<deploy_dir>/ - 若模型文件在本地,SCP 上传:
scp -P <port> <local_model_path> <user>@<ip>:<deploy_dir>/model.hbm - 通过 SSH 验证部署:
ssh -p <port> <user>@<ip> "ls <deploy_dir>/aarch64/bin/hrt_model_exec && ls <deploy_dir>/aarch64/lib/" - 运行
model_info检查模型信息,确认编译核数:
若用户配置了多核 core_id 但模型仅编译为单核,给出警告提示。ssh -p <port> <user>@<ip> "cd <deploy_dir>/script && \ export LD_LIBRARY_PATH=../aarch64/lib/:\$LD_LIBRARY_PATH && \ ../aarch64/bin/hrt_model_exec model_info --model_file=<model_path>"
What ships with it
2 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.
- 11d ago First seen · 286 lines · 70 tokens per session scan E 423942dccc97
j6-ucp-model-perf-eval is a skill published in the GitHub repository HorizonRobotics/OE-Skills (19 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 70 tokens to every session and 3,556 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
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.