ClawHub
Back to skill

Security audit

EASA Regulatory Search

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate EASA regulatory search skill, with disclosed public-source ingestion commands and no evidence of hidden credential use, exfiltration, or destructive behavior.

Install this only if you trust or have reviewed the external `clawEASA` runtime repository and its bootstrap scripts. Use lookup/search commands normally, and run fetch, parse, or FAQ crawling only when you intentionally want to download and update the local public EASA corpus; be careful with the documented `rsync --delete` install command and verify the destination path first.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

High
Confidence
96% confidence
Finding
The skill description promises local querying of a local index, but the documented command set includes remote discovery, downloading, crawling, and ingestion from the EASA website. This mismatch expands the trust boundary from offline/local-only retrieval to network-enabled content acquisition, which can lead to unexpected external access, data exfiltration opportunities, unreviewed content ingestion, and unsafe execution in restricted environments.

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
Website crawling and corpus-ingestion features are broader than the stated purpose of answering regulatory questions from a local index. In the context of an agent skill, these capabilities can trigger unneeded network activity, ingest untrusted remote content into the local corpus, and create persistence or supply-chain risk if the fetched material is later treated as authoritative.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documentation shows an `rsync -a --delete` install command without clearly warning that `--delete` removes files in the destination that are not present in the source. In a skill-install context, users may copy-paste this into a real OpenClaw workspace and unintentionally erase existing files under the target skill directory, especially if the destination path is mistyped or repurposed.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal