ClawHub
Back to skill

Security audit

Agent Cost Strategy

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only cost-optimization skill that may broadly influence model and session choices, but it does not install code, hide behavior, or request unusual privileges.

Use this skill when cost is a priority, but do not let cheap-model defaults override accuracy, safety, or privacy needs. Review provider pricing and cache behavior before relying on the numbers, and end sessions that contain secrets or sensitive personal data even if that costs more.

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

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The description and trigger conditions are extremely broad, causing this skill to activate for many common situations such as general delegation, cost questions, cron jobs, or routine model selection. That increases the chance the skill will steer agent behavior in contexts where cost optimization is not the primary goal, potentially biasing decisions toward cheaper models and longer-lived sessions in ways that may conflict with safety, privacy, or user intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
67% confidence
Finding

The text specifically frames cache behavior around "Anthropic's prompt cache" as the operating assumption, while the skill also claims to work with any AI provider. Although this is not a direct language lock, it introduces a provider-specific operational policy without clear opt-in or documented constraint, which may conflict with the stated cross-provider scope.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.