Back to skill

Security audit

First Aid

Security checks for vulnerabilities and agentic risk

Overview

This is a simple first-aid reference skill with no network access, credential use, hidden execution, or privileged behavior, though its medical guidance should not be treated as authoritative emergency advice.

Install only as a lightweight reference aid. Do not rely on it during a real emergency without contacting local emergency services, and be aware it creates ~/.local/share/first-aid when run.

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 (3)

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
80% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · scripts/script.sh (reported line 9)May include surrounding context.

sh
DATA_DIR="$HOME/.local/share/first-aid"
mkdir -p "$DATA_DIR"

#
#
#
#

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The script presents multiple hard-coded emergency numbers as authoritative guidance without determining the user's locale or clearly warning that the correct number depends on country and service availability. In a first-aid skill, inaccurate emergency-contact guidance can delay contacting responders during an actual emergency, creating real-world safety risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file documents that the skill stores data in ~/.local/share/first-aid/, which affects user data on disk. Under the markdown criteria for missing user warnings, the description should clearly disclose what is stored and whether users can clear or avoid that storage, but it currently provides only a bare location.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.