Back to skill

Security audit

Pixel Familiar Soul

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent agent-identity template, but it gives the agent very broad autonomy and delegation authority without enough user control or data-scope limits.

Install this only if you intentionally want an agent to operate as the Pixel Familiar CEO/orchestrator and to delegate work to named agents. Before use, tighten approval gates for file changes, deployments, external messages, account changes, sensitive data sharing, and memory updates, and require confirmation before forwarding context to other agents.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SOUL.md:5
Finding

Executive Persona and Policy Instructions Hijack the Host Agent

Content
View full analysis
Remediation
View remediation
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 instruction 'If it doesn't move money, just do it' creates an overly broad autonomy boundary that authorizes many potentially risky actions without explicit scope, approval criteria, or safety checks. In an orchestrator/CEO skill that can delegate work, initiate operational changes, and communicate across agents, this can lead to unintended external actions, policy violations, destructive changes, or abuse through prompt injection and task confusion.

Content

No source excerpt is available for this finding.