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 llmcompression-add-modelgit 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/llmcompression-add-model)<a href="https://agentmods.dev/skills/horizonrobotics/oe-skills/llmcompression-add-model"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/llmcompression-add-model/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/llmcompression-add-model"><img src="https://agentmods.dev/badge/skills/horizonrobotics/oe-skills/llmcompression-add-model.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 | $0.00041 | $0.03873 |
| Opus 5 | $0.00020 | $0.01937 |
| Sonnet 5 | $0.00008 | $0.00775 |
| Haiku 4.5 | $0.00004 | $0.00387 |
Grade A, and why
llmcompression-add-model 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 10d 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 — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.
新增 LLM/VLM 模型到 llm_compression
整体流程
- 确定参考源 → leap_llm 或 transformers
- 差异分析 → 详细对比新增模型与参考模型的 diff
- 创建目录结构 →
llm_compression/models/<model_name>/ - 编写 blocks/、
model.py、process_utils.py、<model_name>_model.py - 创建配置文件 →
llm_compression/configs/<model_name>.yml - 注册模型 →
llm_compression/models/__init__.py - 自测验证 → 至少完成 smoke test;如真实权重不可用,做半真实配置验证
- 生成报告 → 输出接入报告并记录关键问题、解决方式和验证结果
Step 0:确定参考源
首先检查 leap_llm/models/ 下有没有同名目录:
- 有:以
leap_llm该模型的forward()方法为主要参考(不是build()) - 没有:去 transformers 源码查找:
<site-packages>/transformers/models/<model_name>/modeling_<model_name>.py
同时参考已集成模型的框架写法:
- VLM:
llm_compression/models/qwen2_5_vl/(无 QK-Norm)或qwen3_vl/(有 QK-Norm + DeepStack) - LLM:
llm_compression/models/qwen3/
当 leap_llm 没有参考模型时(从 transformers 对齐)
必须直接阅读 transformers 源码,提取模型结构信息。详细的阅读方法、config 字段速查和对齐检查清单见 transformers_alignment_guide.md。
核心步骤:
- 阅读
configuration_<model>.py:获取 config 字段及默认值 - 阅读
modeling_<model>.py:提取 Attention/MLP/DecoderLayer/Model/ForCausalLM 结构 - 与框架内参考模型逐项对比差异
Step 0.5:差异分析(必须在写代码前完成)
整理差异清单,包含三类:
- 模型层 diff:attention 形式(q_norm/k_norm、attention_bias、sliding_window)、mlp(bias、激活函数)、norm 类型、rope 实现、权重命名
- 数据层 diff:position_ids 方式(1D 还是 3D mRoPE)、mask 形状、padding 方向、多模态输入
- generate 逻辑层 diff:prefill/decode 流程、chunk_prefill、cache 管理、停止条件
要求:
- 明确指出哪里相同、哪里不同,不要笼统说"类似 qwen3"
- 无法确认的点标记"待确认"
Step 1:blocks/attention.py
1.0 License 头声明(blocks/ 下所有文件 + model.py 必须添加)
models/*/blocks/ 下的所有 .py 文件以及 models/*/model.py 派生自 HuggingFace Transformers(Apache 2.0),必须保留 transformers 原始的完整 License 声明,并追加 Horizon 修改声明。
获取正确的 Copyright 行:查看 transformers/models/<model>/modeling_<model>.py 文件头的 Copyright 行,原样复制。不同模型的 Copyright 行不同(如 Qwen 系列含 Alibaba Group,Gemma 系列只有 HuggingFace Team)。
# Copyright <year> <original authors from transformers>. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# Modifications Copyright (c) Horizon Robotics. All rights reserved.
What ships with it
3 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.
- 10d ago First seen · 306 lines · 41 tokens per session scan A af51b8dc0329
llmcompression-add-model is a skill published in the GitHub repository HorizonRobotics/OE-Skills (19 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 41 tokens to every session and 3,873 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 2026-08-30.
Other skills, from other repositories
agent-platform-rag-engine-management
Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…
training-check
Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.