Back to skill
Skillv1.0.0
ClawScan security
ArXiv Scholar Search · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 9, 2026, 7:11 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's stated purpose (arXiv paper search) matches its instructions and there are no extra installs, credentials, or surprising behaviors requested.
- Guidance
- This skill is coherent and focused: it only documents how to query the public arXiv API and how to format results. Before installing, confirm your agent runtime will (1) enforce the stated rate limit (≤1 request per 3 seconds and one active connection) to avoid API blocks, (2) not augment the skill with unrelated network or file access, and (3) actually source links from the API responses rather than fabricating them. Because the skill origin is unknown, consider running it in a restricted environment or reviewing a sample run to ensure the agent's network calls and outputs match expectations.
Review Dimensions
- Purpose & Capability
- okName and description match the SKILL.md. All required actions and parameters relate directly to querying the arXiv API and formatting results for review.
- Instruction Scope
- okThe SKILL.md limits data sources to arXiv, specifies exact query construction, rate limits, and strict rules against fabricating links or metadata. It does reference a generic 'arxiv_search' call/tool; if your agent/platform lacks that helper it will need to perform standard HTTPS GETs (curl/fetch) consistent with the guide. No instructions ask the agent to read unrelated files, env vars, or to exfiltrate data.
- Install Mechanism
- okThis is an instruction-only skill with no install spec and no code to write to disk, which minimizes install-time risk.
- Credentials
- okNo environment variables, credentials, or config paths are requested. The skill's needs are proportional to an arXiv search workflow.
- Persistence & Privilege
- okalways is false and there is no mechanism to persist or elevate privileges. The skill does not request modification of other skills or system-wide settings.