ClawHub
Back to skill

Security audit

Minimalist Gain

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small reporting helper that runs a local project script and clearly discloses that its report may include a persisted cross-session project ledger.

Install this if you want a project-level minimalist savings report. Be aware that when the repo's reporting script and ledger exist, the output may include previously logged rejected-scope entries from earlier sessions in the same project, so review the ledger contents if that history could contain private or unrelated details.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill advertises reporting for 'this session or project' but directs the agent to read from a persisted cross-session ledger containing data 'logged across sessions, not just this one.' That creates a data-isolation and privacy boundary violation: a user asking about the current session could receive information derived from prior sessions or unrelated work, causing unintended disclosure and misleading attribution.

Content

No source excerpt is available for this finding.