hugging-science

ISSUES FOUND max severity HIGHSkillSpector Caution 22 registry K-Dense-AI last scanned 2026-10-21 1 scan

5 findings (2 high, 2 medium, 1 info), mainly command injection, data exfiltration, policy violation. Most recent scan 2026-10-21.

Findings5

Critical0

High2

Medium2

Low0

Analyzersstatic bytecode pipeline correlation behavioral trigger

Scanner version2.3.1

Second scanner: NVIDIA SkillSpector

VerdictCaution 22

Score22 / 100

Issues2

CategoriesPrivilege Escalation MCP Least Privilege

Version2.12.0 (static)

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

Discovers and evaluates scientific datasets, models, methodology posts, and Spaces through the Hugging Science catalog. Used when selecting scientific ML resources in biology, chemistry, genomics, materials, climate, physics, astronomy, medicine, mathematics, protein design, single-cell analysis, or PDE modeling, and when checking their actual datasets, Transformers, native-runtime, Inference Providers, or Gradio interfaces.

Package: 8 files, 59 KB
SKILL.md: 1548 words, 12+ sections
Code: Python (1)
Compatibility: Requires Python 3.10+ and network access for the standard-library catalog fetcher. Resource use needs the matching scientific runtime; optional HF_TOKEN for gated access.

SKILL.md covers: Hugging Science · When to use this skill · Core workflow · 1. Identify the domain(s) · 2. Fetch the relevant catalog content · 3. Pick the right resource(s) · 4. Use the resource · 5. Cite the methodology · Authentication: HF_TOKEN · A few important things to remember · Review scope · Bundled resources

Findings

HIGH Active instruction acquires and executes remote content Command Injectionstatic

SKILL.md contains an operative statement that explicitly coordinates remote content acquisition with execution.

SKILL.md:136

Reviewed on 2026-10-01 against the public catalog, current author cards and released SDK source. Catalog fetches and public metadata queries ran live; tiny random ESM2, synthetic dataset and mocked cl

Remediation: Remove remote execution, or require a locally pinned artifact whose digest or signature is verified before a fixed allowlisted invocation.

HIGH Python code block uses eval/exec Command Injectionbehavioral

Code block in references/using-models.md at line 42 contains potentially dangerous Python code.

references/using-models.md:42

model = AutoModel.from_pretrained(model_id, revision=revision).eval()

Remediation: Review the code block for security implications.

MEDIUM Outbound network request primitives that can transmit data externally Data Exfiltrationstatic

Pattern detected: urllib.request.Request(

scripts/fetch_catalog.py:97

req = urllib.request.Request(url, headers={"User-Agent": "hugging-science-skill/1.0"})

Remediation: Ensure network operations are necessary and document allowed destinations

MEDIUM Outbound network request primitives that can transmit data externally Data Exfiltrationstatic

Pattern detected: urllib.request.urlopen(

scripts/fetch_catalog.py:99

with urllib.request.urlopen(req, timeout=30) as resp:

Remediation: Ensure network operations are necessary and document allowed destinations

INFO Skill does not specify a license Policy Violationstatic

Skill manifest does not include a 'license' field. Specifying a license helps users understand usage terms.

SKILL.md

Remediation: Add 'license' field to SKILL.md frontmatter (e.g., MIT, Apache-2.0)

What is in the package

File	Kind	Size
`SKILL.md`	doc	12 KB
`references/flagship-resources.md`	doc	6.0 KB
`references/review.md`	doc	5.8 KB
`references/topics-and-slugs.md`	doc	4.5 KB
`references/using-datasets.md`	doc	5.1 KB
`references/using-models.md`	doc	6.7 KB
`references/using-spaces.md`	doc	5.2 KB
`scripts/fetch_catalog.py`	script	14 KB

External hosts referenced

huggingface.co huggingscience.co github.com 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_TOKEN

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-hugging-science-scan.json
Badge for your README: ![skillscan.io](https://skillscan.io/badge/k-dense-ai-hugging-science.svg)
Request a re-scan