Applies to adapting PyTorch-framework VLA models to Ascend NPU, fixing CUDA issues, or optimizing NPU performance. Includes scanning, patching, verification, and training script templates.
- Stars
- 59
- Version
- v1
- Copies
- 0
Overview
Waxell rates it high risk: 8 findings, highest severity high.
- Highest severity
- high
- Findings
- 8
- What ran
- Static analysis + deep scan
A clean result means nothing alarming in what is published, not a guarantee. How the checks work
We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.
Security
Security scan
Checked yesterday
Result
High risk Waxell rates it high risk: 8 findings, highest severity high.
Checked with static analysis + deep scan: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. The deep scan also read the instructions for prompt injection and data exfiltration. Nothing was run. How the checks work
Findings (8)
high · 2
- high
Privilege Escalation: Credential Access
scripts/config.py:21 · skillspector:PE3
- high
YARA Match: YARA rule 'backdoor_persistence': Backdoor persistence with malicious payloads (shell commands, SSH key inje
templates/env_npu.sh:36 · skillspector:YR1
medium · 3
- medium
Without declared permissions the skill's intent is opaque and cannot be validated.
SKILL.md:1 · skillspector:LP3
- medium
Rogue Agent: Session Persistence
templates/train.sh:83 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
templates/train.sh:94 · skillspector:RA2
low · 3
- low
Dangerous Code Execution: Dynamic attribute access via getattr()
scripts/checks/build.py:48 · skillspector:AST7
- low
Dangerous Code Execution: Dynamic attribute access via getattr()
scripts/runtime_patches.py:320 · skillspector:AST7
- low
Dangerous Code Execution: Dynamic attribute access via getattr()
scripts/scanners/__init__.py:65 · skillspector:AST7
Versions
Version history
Each distinct content we have seen, newest first, with the verdict it got at the time.
One version so far: v1, seen 2026-11-16.
See and control which skills your people use
Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.
30 minutes. No commitment.
SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF
More documents skills and plugins
huawei-cloud-skill-audithuaweicloudAudit Huawei Cloud skills for quality, security, and compliance using a three-check pipeline: skillspector (AI security), gitleaks… High riskDocumentsLegal & compliance59
huawei-cloud-vod-collectorhuaweicloudInvoke this skill to capture poor experiences and distill them into high-value requirements (Voice of Developer). Use when user encounters… High riskDocumentsEngineering59
pdf-convertertanis90PDF converter powered by MinerU: convert PDF to Word, Markdown, HTML, LaTeX, or plain text. Also handles image-to-text OCR, scanned… High riskDocumentsWriting & content54
gwskv0906This skill should be used when the user asks to "set up gws", "install Google Workspace CLI", "connect Gmail to Claude", "manage Google… Review findingsDocumentsProductivity52