Security audit

excel-toolkit

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local Excel workbook helper whose file-reading and file-writing behavior matches its stated spreadsheet purpose, with dependency hygiene and documentation clarity caveats.

Install in an isolated Python environment and pin current patched versions of openpyxl and pandas where possible. Use backups or save to new output files before deleting sheets, merging workbooks, or processing spreadsheets from untrusted sources.

SkillSpector

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: The README for an Excel-focused skill advertises unrelated capabilities such as workflow execution, database operations, chart generation, and Feishu Sheets integration. In an agent skill context, this can mislead orchestration or users into invoking broader functionality than declared, increasing the attack surface and creating opportunities for unauthorized data access or unintended tool routing.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The README prominently documents destructive and persistent operations such as writing cells, deleting sheets, saving files, converting formats, and merging workbooks without warning about overwriting or modifying user data. In an agent setting, this raises the risk of unsafe autonomous file edits because users and higher-level systems are not prompted to use backups, confirmation steps, or output-to-new-file defaults.

Unpinned Dependencies

Low

Category: Supply Chain
Content: openpyxl pandas
Confidence: 94% confidence
Finding: The dependency is unpinned, so installs may resolve to different versions over time, including versions with newly introduced vulnerabilities or breaking behavior. In a file-processing skill that handles Excel content, this increases supply-chain and reproducibility risk because parsing libraries are exposed to attacker-controlled files.

Unpinned Dependencies

Low

Category: Supply Chain
Content: openpyxl pandas
Confidence: 94% confidence
Finding: The pandas dependency is also unpinned, allowing uncontrolled version drift during installation. That can introduce vulnerable or incompatible versions into a skill that processes user-supplied spreadsheet or tabular data, making exploitation and reliability issues more likely.

Known Vulnerable Dependency: openpyxl — 2 advisory(ies): CVE-2017-5992 (Improper Restriction of XML External Entity Reference in Openpyxl); CVE-2017-5992 (Openpyxl 2.4.1 resolves external entities by default, which allows remote attack)

High

Category: Supply Chain
Confidence: 83% confidence
Finding: The requirement references openpyxl without a version bound, which permits installation of versions affected by CVE-2017-5992 involving XML external entity processing. Because this skill is explicitly designed to inspect and edit Excel/XLSX files, it is more dangerous than usual: attacker-provided workbook content could reach the vulnerable parser path and potentially enable XXE-style file disclosure or network access depending on runtime configuration.

Static analysis

No suspicious patterns detected.