transformers
1 finding (1 high), mainly command injection. Most recent scan 2026-10-21.
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Hugging Face Transformers for loading Hub models, running pipeline inference, text generation, and Trainer fine-tuning on NLP, vision, audio, and multimodal tasks. Applies when working with AutoModel, pipelines, tokenizers, generation configs, or TrainingArguments within Transformers.
- Package
- 7 files, 79 KB
- SKILL.md
- 1264 words, 12+ sections
- Code
- None. Instructions only.
- Allowed tools
ReadWriteEditBash- License
- Apache-2.0 license
- Compatibility
- Requires Python 3.10+, PyTorch 2.5+, and transformers 5.18.0; optional librosa 1.0 requires Python 3.12+. Network access for Hub downloads. Gated or private Hub models need an HF token (`hf auth login` or `HF_TOKEN`).
SKILL.md covers: Transformers · Overview · Installation · Authentication · Transformers v5 · Quick Start · Text generation (prefer max_new_tokens for causal LMs) · Text classification · Generative question answering: verify responses against the supplied context. · Core Capabilities · 1. Pipelines for Quick Inference · 2. Model Loading and Management
Findings
HIGH Python code block uses eval/exec Command Injectionbehavioral
Code block in references/models.md at line 290 contains potentially dangerous Python code.
references/models.md:290
model.eval()What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 11 KB |
references/generation.md | doc | 11 KB |
references/models.md | doc | 11 KB |
references/pipelines.md | doc | 12 KB |
references/review.md | doc | 8.1 KB |
references/tokenizers.md | doc | 11 KB |
references/training.md | doc | 15 KB |
External hosts referenced
github.com huggingface.co pypi.org arxiv.org doi.org export.arxiv.org
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Environment variables referenced
HF_HOME HF_TOKEN HF_HUB_CACHE HF_TOKEN_PATH
Package contents recorded 2026-10-22 at commit 593377bb0487. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Bytecode analyzer.
- Pipeline analyzer.
- Correlation analyzer.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://github.com/K-Dense-AI/scientific-agent-skills
- Raw scan result: skills-transformers-scan.json
- Badge for your README:
- Request a re-scan