Trust Assessment
pdf received a trust score of 90/100, placing it in the Trusted category. This skill has passed all critical security checks and demonstrates strong security practices.
SkillShield's automated analysis identified 2 findings: 0 critical, 0 high, 2 medium, and 0 low severity. Key findings include Command-line examples vulnerable to injection with untrusted filenames, Skill allows writing to arbitrary file system locations.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Command-line examples vulnerable to injection with untrusted filenames The skill provides examples of using command-line tools like `pdftotext`, `qpdf`, `pdftk`, and `pdfimages` within `SKILL.md`. If the AI agent uses these examples with filenames derived from untrusted user input without proper sanitization or quoting, it could lead to command injection. For instance, a filename like `my_doc.pdf; rm -rf /` could execute arbitrary commands. The skill does not provide guidance on sanitizing filenames for these shell commands, which could lead to the agent constructing vulnerable commands. Add explicit warnings and instructions for the AI agent to sanitize or properly quote filenames when constructing shell commands based on user input. Recommend using Python's `subprocess.run` with `shell=False` and a list of arguments for robustness, or a dedicated shell escaping utility. | LLM | SKILL.md:160 | |
| MEDIUM | Skill allows writing to arbitrary file system locations The skill's Python scripts (e.g., `scripts/convert_pdf_to_images.py`, `scripts/create_validation_image.py`, `scripts/extract_form_field_info.py`, `scripts/fill_fillable_fields.py`, `scripts/fill_pdf_form_with_annotations.py`) are designed to read from and write to file paths provided as arguments via `sys.argv`. This capability, while necessary for PDF processing, allows the AI agent to write to any location on the file system where it has permissions. If the agent is prompted to write to sensitive system directories or user-specific configuration files, it could lead to data corruption, privilege escalation, or other security breaches. The skill does not include mechanisms to restrict output paths to a safe sandbox. Implement sandboxing or strict path validation within the scripts to ensure that output files can only be written to designated safe directories (e.g., a temporary directory, a user-specific output folder). Provide clear guidance to the AI agent on safe file handling practices. | LLM | SKILL.md |
Scan History
Embed Code
[](https://skillshield.io/report/3e4f990b6d5221d6)
Powered by SkillShield