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-infer-generatinggit 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-infer-generating)<a href="https://agentmods.dev/skills/horizonrobotics/oe-skills/j6-ucp-infer-generating"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-infer-generating/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-infer-generating"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-ucp-infer-generating.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 | $0.00086 | $0.00675 |
| Opus 5 | $0.00043 | $0.00338 |
| Sonnet 5 | $0.00017 | $0.00135 |
| Haiku 4.5 | $0.00009 | $0.00068 |
Grade A, and why
j6-ucp-infer-generating 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.
What it actually says
基本推理步骤
- Step 01 - .hbm 模型加载
references/steps/step-01-load-model.md - Step 02 - 张量信息与内存申请
references/steps/step-02-prepare-tensor.md - Step 03 - 预处理（数据填充与缓存刷新）
references/steps/step-03-preprocess.md - Step 04 - 执行推理
references/steps/step-04-run-inference.md - Step 05 - 输出解析
references/steps/step-05-postprocess.md - Step 06 - 资源释放
references/steps/step-06-release-resource.md
文档路径均为本目录下的相对路径。其中 Step 03/04/05 需要每个推理帧重做，其他步骤可以复用。
分析行为规则
- 当用户的生成要求涉及到一个或多个步骤时，再阅读对应的步骤文档。
- 遇到不明确的 API 调用时，阅读
references/api/index.md路由详细 API 介绍。
代码生成规则
- 代码生成聚焦 UCP 接口调用，无关代码（信息打印、非 UCP 接口返回值的检查、异常处理、命令行参数解析等）默认不关注。
- 代码生成时必须参考用户代码上下文，统一代码风格，复用已有逻辑。
代码生成流程
必须逐个 step 生成代码，用户确认后再进入下个 step 分析生成流程
- 阅读用户代码上下文和用户请求。
- 确定代码生成涉及的 Steps 列表。
- 对每个涉及的 Step 串行执行: 3.1 阅读对应的 Step 文档。 3.2 根据上下文或用户请求确定具体分支，若可以确定 -> 继续下一步；否则 -> AskUserQuestion -> 重新执行此步骤。 3.3 确定要使用的 UCP API 列表。 3.4 对每个涉及的 API 串行执行: 3.4.1 确定 API 所在头文件。 3.4.2 根据上下文或用户请求确定 API 的每个参数，若可以确定 -> 继续下一步；否则 -> AskUserQuestion -> 重新执行此步骤。 3.4.3 API 参数均确定后进行下一个 API 的处理。 3.5 分支均确定后进行下一个 Step 的处理。
- 执行代码生成并检查正确性。
What ships with it
52 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.
- CHANGELOG.md 520 B
- example.md 15 KB
- references/api/dnn/hbDNNDataType.md 717 B
- references/api/dnn/hbDNNDescType.md 266 B
- references/api/dnn/hbDNNGetCompileBpuCoreNum.md 234 B
- references/api/dnn/hbDNNGetHBMDesc.md 391 B
- references/api/dnn/hbDNNGetInputCount.md 204 B
- references/api/dnn/hbDNNGetInputDesc.md 251 B
- references/api/dnn/hbDNNGetInputName.md 223 B
- references/api/dnn/hbDNNGetInputTensorProperties.md 340 B
- references/api/dnn/hbDNNGetModelDesc.md 228 B
- references/api/dnn/hbDNNGetModelHandle.md 372 B
- references/api/dnn/hbDNNGetModelNameList.md 350 B
- references/api/dnn/hbDNNGetOutputCount.md 207 B
- references/api/dnn/hbDNNGetOutputDesc.md 254 B
- references/api/dnn/hbDNNGetOutputName.md 226 B
- references/api/dnn/hbDNNGetOutputTensorProperties.md 263 B
- references/api/dnn/hbDNNGetTaskOutputTensorProperties.md 443 B
- references/api/dnn/hbDNNGetVersion.md 232 B
- references/api/dnn/hbDNNInferV2.md 847 B
- references/api/dnn/hbDNNInitializeFromDDR.md 385 B
- references/api/dnn/hbDNNInitializeFromFiles.md 362 B
- references/api/dnn/hbDNNQuantiScale.md 479 B
- references/api/dnn/hbDNNQuantiType.md 285 B
- references/api/dnn/hbDNNRelease.md 260 B
- references/api/dnn/hbDNNTensor.md 311 B
- references/api/dnn/hbDNNTensorProperties.md 626 B
- references/api/dnn/hbDNNTensorShape.md 313 B
- references/api/index.md 1.4 KB
- references/api/ucp/hbUCPFlushLog.md 148 B
- references/api/ucp/hbUCPFree.md 160 B
- references/api/ucp/hbUCPGetErrorDesc.md 199 B
- references/api/ucp/hbUCPGetSocName.md 153 B
- references/api/ucp/hbUCPGetVersion.md 154 B
- references/api/ucp/hbUCPMalloc.md 183 B
- references/api/ucp/hbUCPMallocCached.md 204 B
- references/api/ucp/hbUCPMemFlush.md 186 B
- references/api/ucp/hbUCPReleaseTask.md 177 B
- references/api/ucp/hbUCPSchedParam.md 319 B
- references/api/ucp/hbUCPSetTaskDoneCb.md 222 B
- references/api/ucp/hbUCPStatus.md 1.7 KB
- references/api/ucp/hbUCPSubmitTask.md 214 B
- references/api/ucp/hbUCPSysMem.md 250 B
- references/api/ucp/hbUCPSysMemFlushFlag.md 201 B
- references/api/ucp/hbUCPTaskPriority.md 194 B
- references/api/ucp/hbUCPWaitTaskDone.md 193 B
- references/steps/step-01-load-model.md 2.0 KB
- references/steps/step-02-prepare-tensor.md 5.1 KB
- references/steps/step-03-preprocess.md 1.0 KB
- references/steps/step-04-run-inference.md 3.7 KB
- references/steps/step-05-postprocess.md 1.1 KB
- references/steps/step-06-release-resource.md 836 B
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 · 49 lines · 86 tokens per session scan A 89268e8e2806
j6-ucp-infer-generating is a skill published in the GitHub repository HorizonRobotics/OE-Skills (19 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 86 tokens to every session and 675 once invoked, about $0.0004 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
ax-cpp-ai
Use when writing C++ code with axllm for named deployment profiles, generic provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers.
ax-cpp-gen
Use when writing C++ code with axllm for AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing.
ax-cpp-llm
Use when writing C++ code with axllm for using the generated Ax package, factory functions, package docs, examples, and API reference.
ax-cpp-signature
Use when writing C++ code with axllm for string signatures, field descriptors, JSON schema output, validation, and typed tool argument shapes.
ax-cpp-gepa
Use when writing C++ code with axllm for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.
cuda-kernels
Provides guidance for writing and benchmarking optimized CUDA kernels for NVIDIA GPUs (H100, A100, T4) targeting HuggingFace diffusers and transformers libraries. Kernels must be kernel-builder/ABI3-compliant: no pybind11, no setup.py, TORCHLIBRARYEXPAND bindings only. Supports models like LTX-Video, Stable Diffusion…