ClawHub

paper-review

Security checks across malware telemetry and agentic risk

Overview

This is a coherent academic paper-review helper, but users should be mindful that papers and reference data may be processed by external model and bibliography services.

Install only if you trust the OpenJudge/LiteLLM package chain and the model provider or custom endpoint you configure. Prefer environment variables or restricted API keys, avoid echoing secrets in terminals, and do not submit confidential, embargoed, or unpublished manuscripts to external services unless that use is approved.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill description is broad enough to trigger on generic requests to 'review, evaluate, critique, or assess a research paper' and also to 'check references' or 'verify a BibTeX file,' which can overlap with more general document-analysis tasks. This can cause the agent to invoke a heavyweight external-review pipeline in contexts where the user did not clearly ask for this specific capability, increasing the chance of unintended file handling, external API use, and disclosure of user-supplied content to third-party services.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The reference explicitly supports sending paper content, metadata, and bibliographic details to external model providers and APIs through configurable `model_name`, `api_key`, `base_url`, and CrossRef integration, but it does not clearly warn users that potentially unpublished or sensitive research materials may leave the local environment. In a paper-review skill, this matters because users may submit confidential manuscripts, reviewer comments, or embargoed work, creating privacy, confidentiality, and policy-compliance risks if data is transmitted to third parties without informed consent.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal