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-plugin-preparegit 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-plugin-prepare)<a href="https://agentmods.dev/skills/horizonrobotics/oe-skills/j6-plugin-prepare"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-plugin-prepare/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-plugin-prepare"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/j6-plugin-prepare.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 | $0.00050 | $0.01251 |
| Opus 5.5 | $0.00020 | $0.00500 |
| Sonnet 5.5 | $0.00010 | $0.00250 |
| Haiku 4.5 | $0.00005 | $0.00125 |
Grade A, and why
j6-plugin-prepare 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 2mo 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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
给浮点模型添加 Horizon prepare(仅 prepare 调用版)
目标
把已有的 浮点 torch.nn.Module 接入 horizon_plugin_pytorch 的 QAT 工具链:在代码中对模型执行 prepare(...),得到可训练的 QAT 模型(含伪量化/qat 算子等变换)。
本 Skill 强约束:
- 只加 prepare:仅新增/调整
prepare(...)调用与必要 import,不做其它结构性改造。 - 不处理 dynamic_block:不引入/不修改任何
dynamic_block/Tracer.dynamic_block相关逻辑(动态控制流场景需要单独处理,不属于本 skill)。 - qconfig_setter 固定:一律使用“全部双 int8”模板构造的
QconfigSetter,不叠加其它 setter,不做敏感算子表等策略。
prepare 会做什么(你需要知道的关键点)
基于图的 prepare(例如 PrepareMethod.JIT_STRIP)通常会:
- 捕获并裁剪计算图(会根据
QuantStub/DeQuantStub位置识别并跳过前后处理) - 替换部分 function 类算子为 module 形式(便于在 module 内插入伪量化等逻辑)
- 进行可融合 pattern 的算子融合
- 将浮点算子转换为 qat 算子,并按 qconfig 插入伪量化/伪转换节点
- 执行 QAT 模型结构检查并生成检查结果文件
标准改法(通用模板)
1) 增加 import
from horizon_plugin_pytorch.dtype import qint8
from horizon_plugin_pytorch.quantization import get_qconfig
from horizon_plugin_pytorch.quantization.prepare import PrepareMethod, prepare
from horizon_plugin_pytorch.quantization.qconfig_setter import (
ConvDtypeTemplate,
MatmulDtypeTemplate,
ModuleNameTemplate,
QconfigSetter,
)
2) 选择 method,并准备 example_inputs
- 推荐默认用
PrepareMethod.JIT_STRIP - 当 method 为
PrepareMethod.JIT_STRIP或PrepareMethod.JIT时,必须提供example_inputs example_inputs的目的:用于感知图结构,且应当能跑通目标 forward(通常用 eval/infer 路径)
3) 构造全双 int8 的 qconfig_setter,并调用 prepare
int8_qconfig_setter = QconfigSetter(
reference_qconfig=get_qconfig(),
templates=[
ModuleNameTemplate({"": qint8}),
ConvDtypeTemplate(input_dtype=qint8, weight_dtype=qint8),
MatmulDtypeTemplate(input_dtypes=qint8),
],
)
qat_model = prepare(
float_model,
example_inputs=example_inputs,
qconfig_setter=int8_qconfig_setter,
method=PrepareMethod.JIT_STRIP,
)
4) 重要约束与注意事项(直接按这个执行)
- prepare 之后不要再改模型结构:prepare 会替换/融合/转换算子,之后再改模型(例如把 BN 改成 SyncBN)可能导致 qat 算子被再次修改,出现不可预期行为。
- prepare 之后不要改任何 hook:基于图的 function→module 替换依赖 hook 与特殊 wrapper tensor 机制;改动 hook 可能导致替换失效,进而报错或产生精度问题。
- 尽量只对“部署逻辑”做 prepare:如果一个
forward同时混入训练/评测分支、CPU/Numpy 后处理等,容易导致量化边界或图捕获不符合部署预期。建议把部署推理逻辑剥离到forward_infer(或等价函数)并对其进行 prepare。
What ships with it
1 file 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.
- 2mo ago First seen · 92 lines · 50 tokens per session scan A d92c8c3649b4
j6-plugin-prepare is a skill published in the GitHub repository HorizonRobotics/OE-Skills (27 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 50 tokens to every session and 1,251 once invoked, about $0.0002 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 2026-08-30.
j6-plugin-prepare compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| j6-plugin-prepareHorizonRobotics/OE-Skills | Skill | 27 | 4mo ago | Apache-2.0 | A | 50 | Claude Code, Codex |
| agent-platform-model-registrygoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 60 | Claude Code, Codex |
| foundry-config-setupmicrosoft/agent-framework | Skill | 15k | yesterday | MIT | A | 65 | Claude Code, Codex |
| google-cloud-solution-agentic-analytics-spark-knowledge-cataloggoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 132 | Claude Code, Codex |
| training-checkwanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | 3d ago | MIT | A | 35 | Claude Code |
| nemo-automodel-launcher-configNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 30 | Claude Code |
| finetuningmicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 132 | Claude Code |