Back to skill

Security audit

Freelance Delivery Team — Agent Configuration Bundle

Security checks for vulnerabilities and agentic risk

Overview

This skill clearly describes a supervised freelance-work team, but users should still review its file, shell, memory, inbox, and payment-related access before enabling it.

Install only if you are comfortable with a skill that can help search for freelance work, write proposals, run local work, read and write files, use shell commands, store job context in memory, and review job-related inbox/payment evidence. Keep it limited to dedicated workspaces and approved job/payment sources.

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
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill advertises privileged capabilities including shell execution, file read/write, and inbox/payment monitoring, but provides no explicit user-facing warning, consent boundary, or safety constraints around system changes, data access, or privacy-sensitive handling. In this context, the workflow is designed to autonomously scout jobs, execute work locally, monitor communications, and track payments, which increases the chance of unintended data exposure, unauthorized local changes, or overbroad access if the skill is enabled without clear disclosure and restrictions.

Content

No source excerpt is available for this finding.