Back to skill
Skillv0.1.0
ClawScan security
paper-writing-studio · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 30, 2026, 4:01 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's files, instructions, and requested resources are consistent with a paper-writing workflow for news & communication; there are no unexpected credentials, network endpoints, or risky install steps.
- Guidance
- This skill appears internally consistent and low-risk: it only includes local templates, references, and two small Python helpers for classification and checklist generation. Before using, consider: (1) avoid pasting unpublished sensitive data or full manuscripts into third-party models if you plan to use external LLMs or online search tools the skill recommends; (2) the classification logic is simple rule-based and may misclassify edge cases—verify stage/type before acting; (3) if you plan to run the included scripts, ensure Python is available and review them (they are short and non-networking). If you need the skill to access subscription databases (CNKI, WOS), you will need to provide those credentials separately—this skill does not request or store them.
Review Dimensions
- Purpose & Capability
- okName/description (paper-writing workflow for 新闻与传播) match the included assets (stage playbook, topic ideas, templates) and the two small helper scripts. All declared and required resources are appropriate for this purpose.
- Instruction Scope
- okSKILL.md confines actions to structuring writing stages, recommending tools, reading bundled reference files, and using simple local scripts to classify tasks/generate checklists. It explicitly warns against fabricating citations and not to invent retrieved literature. No instructions request reading unrelated system files or exfiltrating data.
- Install Mechanism
- okNo install spec (instruction-only) and the only code are two short Python scripts and a set of markdown templates/references. Nothing downloads external archives or creates binaries; low disk/execution risk. Scripts are simple CLI utilities without network calls.
- Credentials
- okThe skill does not request environment variables, credentials, or config paths. Recommended external tools/services (CNKI, Google Scholar, large models) are suggested for the user/agent to use but are not required by the skill itself.
- Persistence & Privilege
- okalways is false and the skill does not request persistent system presence or modification of other skills or system-wide settings. Normal user-invocable/autonomous invocation settings are used.