ClawHub
Back to skill

Security audit

OpenClaw Knowledge Base Retriever

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed local knowledge-base retriever with proportionate optional file outputs and no evidence of hidden data sharing or destructive behavior.

Install only if you are comfortable letting the skill read the chosen knowledge-base directory and create local helper files such as data_structure.md indexes, extracted PDF text, or page images. Review prompts before approving pip installs or any --force index rebuild because those change the local Python environment or overwrite existing index files.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill performs file read and file write operations but does not declare permissions, which weakens transparency and reviewability. In practice this can mislead users or orchestrators into invoking a supposedly read-focused skill that can also create or overwrite local files such as data_structure.md indexes, extracted text, or image outputs.

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding
The skill is described primarily as a local knowledge-base retriever, but its documented behavior also includes generating or updating index files and producing derivative artifacts from PDFs. This mismatch is dangerous because users and policy systems may trust it as a read-only retrieval tool while it can modify the filesystem and potentially overwrite existing index files or create large volumes of output data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.