ClawHub
Back to skill

Security audit

Anthropic Cost Optimizer

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent OpenClaw cost optimizer, but its broad trigger wording could cause local config files to be read in conversations that are only loosely related to OpenClaw cost tuning.

Install only if you are comfortable with the agent inspecting OpenClaw config files when cost-optimization prompts are detected. Before using it, make sure the prompt clearly asks for OpenClaw or Anthropic API cost optimization, review the shown diff carefully, and confirm only the changes you actually want written.

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 (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill advertises broad trigger phrases like "save money," "too expensive," billing changes, and generic config-help requests, which can cause the agent to invoke this skill in many ordinary conversations that are only loosely related. Because the skill reads live config files and can rewrite them after a confirmation step, over-triggering increases the chance of unnecessary access to sensitive local configuration and unsafe or undesired config modifications.

Content

No source excerpt is available for this finding.