Security audit

EPUB reader

Security checks for vulnerabilities and agentic risk

Overview

This EPUB skill is mostly coherent, but it asks agents to run unsafe archive extraction and system-level Python package installation patterns that users should review before installing.

Install only if you are comfortable adapting the workflow to a safer setup: use a virtual environment or preinstalled packages, avoid `--break-system-packages`, extract each EPUB into a fresh temporary directory with archive validation and size limits, and choose unique output filenames to avoid overwriting prior work.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (2)

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:27
Finding: Unbounded Extraction of an Untrusted EPUB Archive into a Reused Directory
Content: View full analysis
Remediation: View remediation

T08 · Insecure Dependencies

Warning

Location: SKILL.md:284
Finding: Unpinned Dependency Installation Bypasses System Python Protections
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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 (4)

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill instructs installing packages with pip install ... --break-system-packages, which weakens environment protections and can modify the system-wide Python environment unnecessarily for a document-processing task. In an agent setting, this increases risk of dependency tampering, environment corruption, and unintended side effects beyond the skill's scope.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The example writes extracted text directly to /mnt/user-data/outputs/book_full_text.txt without warning that an existing file may be overwritten. While low severity, silent file creation or overwrite can cause data loss or confusion in automated workflows where output paths are reused.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The cover extraction example copies a file to a fixed output path without noting that it may create or overwrite /mnt/user-data/outputs/cover.jpg. In isolation this is minor, but in repeated runs or shared workflows it can unexpectedly replace prior user data.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 86% confidence
Finding: The repack workflow copies the generated EPUB to a fixed output filename without warning that an existing modified_book.epub may be replaced. This is a real but low-impact safety issue because it can destroy previous output artifacts or mislead users about which file is current.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.