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 dslsdzc/rev-skills --skill re-ai-modelgit clone --depth 1 https://github.com/dslsdzc/rev-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/dslsdzc/rev-skills/re-ai-model)<a href="https://agentmods.dev/skills/dslsdzc/rev-skills/re-ai-model"><img src="https://agentmods.dev/badge/skills/dslsdzc/rev-skills/re-ai-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/dslsdzc/rev-skills/re-ai-model"><img src="https://agentmods.dev/badge/skills/dslsdzc/rev-skills/re-ai-model.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00097 | $0.04160 |
| Opus 5 | $0.00048 | $0.02080 |
| Sonnet 5 | $0.00019 | $0.00832 |
| Haiku 4.5 | $0.00010 | $0.00416 |
Grade A, and why
re-ai-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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI 模型逆向（ONNX / PyTorch / Safetensors）
何时使用 / 何时不用
- 用：拿到 .onnx / .pt / .pth / .safetensors / .tflite 等模型文件，要还原网络结构、提取权重
- 用：文件级水印检测（东西藏在哪里：权重 pattern / metadata / tensor hash / embedding 异常——怀疑模型是从原版复制/微调而来时先查文件侧）
- 用：模型文件本身是载荷——权重里藏数据、torch.save 打包恶意 pickle、后门/投毒模型（下载执行类样本）
- 不用：行为级水印（模型表现出来是什么：trigger 触发响应 / 查询响应 / 黑盒指纹）与 API 行为层攻击（走 [[re-ai-attack]]）
- 不用：纯推理脚本/训练代码（那是源码，走 [[re-script-deob]]）
- 不用：模型被打包进可执行文件（PyInstaller/pyarmor 等）——先 [[re-binary-core]] 拆包，拆出的模型文件再回本技能
- 边界：本技能定位 = 模型文件解析 / 结构分析 / 权重分析 / 文件级水印——恶意模型判定、投毒/后门行为侧、归属取证属取证域（未来独立 re-ai-malware 技能承接；当前此类需求暂在本技能范围，以安全边界（坑 2 pickle 隔离）处理，行为侧转 [[re-ai-attack]]）
- 注意：安全提示——不要直接 torch.load 未知 pkl 文件（pickle 反序列化可执行任意代码，见坑 2）；一切对未知 pkl 的加载默认隔离环境（[[platform-tips]] 沙箱最高原则），先读后跑；模型解析/权重提取为静态步骤，可免沙箱
工具准备
参考 [[platform-tips]]——模型文件 GB 级常见，静态分析按「静态优先（大型样本）」思路：先格式识别与结构解析，按需提取权重，不整载内存（坑 1）。
python3 —— 所有解析脚本基础
- Linux:
apt install python3/dnf install python3/pacman -S python - macOS:
brew install python3；Windows:choco install python - 验证:
python3 --version（本技能脚本均为 Python 3）
onnx（pip，Python 3.10+）—— ONNX 解析主力
pip install onnx（官方 PyPI；onnx 1.22 要求 Python 3.10+，自带 protobuf 依赖与onnx.proto3类型定义）- 验证:
python3 -c "import onnx; print(onnx.__version__)"
netron（pip，Python 3）—— 模型可视化
pip install netron（官方 PyPI，无 Python 版本上界）；桌面独立版可选: macOSbrew install --cask netron、Windowswinget install netron、Linux snapsnap install netron- 验证:
netron --help有输出（pip show netron查版本） - 用法:
netron model.onnx（本地起 http 服务并开浏览器可视化；--no-browser无头模式）
torch（pip，Python 3.9+）—— PyTorch 模型加载
- Linux/Windows:
pip install torch（默认 PyPI 轮子为带 CUDA 全量包，数 GB；仅 CPU 分析用pip install torch --index-url https://download.pytorch.org/whl/cpu） - macOS:
pip install torch（官方 wheel 为 CPU/arm64） - 验证:
python3 -c "import torch; print(torch.__version__)" - 安全注：torch.load 底层是 pickle——不要直接 load 未知 pkl 文件；PyTorch 2.6+ 默认
weights_only=True，旧版本/显式weights_only=False仍有任意代码执行风险；未知模型先unzip -l/xxd粗查（坑 2），在隔离环境用weights_only=True加载，能转 safetensors 就转
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 · 161 lines · 97 tokens per session scan A 747359183e3d
re-ai-model is a skill published in the GitHub repository dslsdzc/rev-skills (52 stars, last pushed 12d ago), licensed Apache-2.0. It adds 97 tokens to every session and 4,160 once invoked, about $0.0005 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
analyzing-golang-malware-with-ghidra
Reverse engineer Go-compiled malware in Ghidra by parsing Go buildinfo and pclntab structures, recovering stripped/obfuscated function names (e.g. via GoResolver), and extracting embedded module/dependency strings and types from Go binaries. Use when analyzing a Go-language malware sample, deobfuscating a…
torchdrug
Build and troubleshoot TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, and knowledge graph reasoning. Use when code imports torchdrug or needs its datasets, models, tasks, or Engine.
deepspot-m
Generate transcriptome-wide virtual spatial transcriptomics from H&E histology with DeepSpot-M. Use when you need spatial gene expression in log1p-CPM for 224x224 tiles at about 20x, want to query protein-coding genes by symbol instead of a fixed panel, or want to run prediction across a whole slide after tiling with…
nemo-mbridge-perf-expert-parallel-overlap
Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlapmoeexpertparallelcomm, delaywgradcompute, and flex dispatcher backends such as DeepEP and HybridEP.
pick-a-pii-model
Select an on-device OpenMed PII model from the committed registry by language, runtime format, and size budget, then require recall validation before deployment. Use when an agent must choose a local PII detector for CPU, Apple Silicon, or a mobile export without relying on live model discovery.
esm
Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel…