Back to skill
Skillv1.0.0
ClawScan security
Research Paper Orchestrator · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 6:30 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- The skill is a prompt-only academic writing workflow with no code, install steps, credentials, persistence, or hidden data-access behavior shown in the artifacts.
- Guidance
- This appears low-risk from a security perspective. Before using it for academic work, independently verify sources, citations, and institutional rules, because the artifacts describe a writing workflow but do not provide a separate mechanism for guaranteeing citation accuracy or academic acceptability.
Review Dimensions
- Purpose & Capability
- okThe described capabilities—planning, literature review, drafting, quality review, and Word-document production—are coherent with the stated research-paper orchestration purpose.
- Instruction Scope
- okThe instructions are user-invoked and ask for topic, citation style, discipline, research type, requirements, and timeline before proceeding; no goal override, forced tool use, or unsafe stopping-condition changes are evident.
- Install Mechanism
- okThere is no install spec, no required binaries, no required environment variables, and no code files, so the artifacts do not show package, script, or executable supply-chain risk.
- Credentials
- okThe artifacts do not request local filesystem access beyond producing a document, credentials, browser/session data, network endpoints, or operating-system privileges.
- Persistence & Privilege
- okNo persistence mechanism, background worker, privileged configuration, or ongoing autonomous activity is described.