Back to skill

Security audit

Payout Possum

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed money-recovery helper whose sensitive data requests fit the purpose, though users should share only the minimum needed.

Before installing, be aware that effective money-recovery searches may involve names, prior addresses, employers, banks, insurers, and possibly Gmail messages. Approve inbox access only when you want that coverage, avoid sharing full account numbers or unnecessary identifiers, and back up any existing local payout-possum customization before running the installer.

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
  • 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 (5)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs collection of a broad set of sensitive personal data including full legal names, prior names, addresses, phone numbers, email addresses, employers, financial institutions, and account history, but provides no privacy minimization, retention, consent, or handling warning. Even if the business purpose is legitimate, aggregating this identity profile in a conversational workflow increases exposure to identity theft, account recovery abuse, and unnecessary overcollection if the data is logged, retained, or later misused.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The default activation prompt is very broad and encourages use for a wide range of money-recovery investigations without clear boundaries, prerequisites, or consent checks. In this skill’s context, that breadth can lead to over-collection of sensitive financial and email data, unnecessary invocation in ambiguous situations, and expansion into higher-risk account-search or claims-assistance behaviors.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly directs Gmail inbox searches and message summarization, including extracting deadlines, claimant IDs, source names, and links, but does not require a clear user-facing warning that sensitive email contents may be read. In this context, the omission creates a real privacy risk because the skill is designed to inspect potentially broad categories of personal financial and legal correspondence, increasing the chance of over-collection or unexpected access to sensitive data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The script unconditionally deletes the existing target skill directory with rm -rf before reinstalling, without prompting, backing up, or validating whether the destination contains local modifications. While the path is not attacker-controlled in this snippet, this behavior can still cause destructive loss of user data or locally customized skill content if the script is run unintentionally or repeatedly.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
85% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · scripts/install-local.sh (reported line 24)May include surrounding context.

sh
fi
}

install_dir "${HOME}/.codex/skills"
install_dir "${HOME}/.openclaw/skills"

echo "Installed payout-possum to:"