ClawHub
Back to skill

Security audit

Vector Embedded Retrieval Archive

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent local document search helper that clearly distinguishes read-only search from user-authorized file-writing operations.

Before installing, understand that this skill can help an agent read and search local document archives. Approve conversion, indexing, overwrite, or export commands only when you intend to create or replace files, and be careful with sensitive PDF libraries because .vera archives and .vera-index folders may persist derived content locally.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • 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)

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · references/cli-reference.md (reported line 492)May include surrounding context.

md
This command exits 0 only when every case in every requested mode hits. A miss
still prints the report and exits 1.

## Exit and output rules

All JSON-capable commands print one JSON object to stdout on success. Check the
exit code before deciding how to interpret output:

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The markdown states that automatic conversion uses bundled English language OCR data and that other languages require external Tesseract data. This is a natural-language locale constraint that defaults behavior to English without presenting a user choice or framing it as a clearly justified region-specific limitation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.