ClawHub
Back to skill

Security audit

skill-retrieval-gate

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small guidance-only memory-retrieval gate that is disclosed, limited, and aligned with its stated purpose.

Installers should understand that this skill may cause an agent to consult local memory when a task appears to depend on prior project context, but it is designed to limit retrieval and avoid using memory for unrelated one-shot tasks.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest description lists trigger phrases such as 'continue previous work', 'check what we already documented', and 'base this on existing notes', which are common phrasings that could appear in many ordinary requests. Although examples are provided, the file does not define clear negative examples or explicit boundaries for when the skill should not activate, increasing the risk of unintended invocation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The 'Example triggers' section provides illustrative requests, but several are phrased broadly enough to overlap with routine collaboration language. The section does not clearly distinguish between cases that should invoke this skill and similar requests that should proceed without it, leaving the activation condition underspecified.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.