Security audit

WMU志愿时长收集器

Security checks for vulnerabilities and agentic risk

Overview

The skill matches its volunteer-hour spreadsheet purpose, but it needs review because it bulk-processes student personal data and can write unsafe spreadsheet formulas into the output workbook.

Install only for the intended university volunteer-hour workflow. Use a dedicated folder containing only the spreadsheets you intend to process, confirm authorization to handle student IDs and phone numbers, and avoid sharing generated workbooks until formula-prefixed values are sanitized. Install dependencies in a virtual environment, preferably from pinned hashes, and note that the referenced template asset is missing from this artifact.

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: scripts/helpers.py:157
Finding: Untrusted Spreadsheet Values Can Be Written as Active Excel Formulas
Content: View full analysis
Remediation: View remediation

T08 · Insecure Dependencies

Note

Location: SKILL.md:12
Finding: Runtime Dependencies Are Installed Without Version or Integrity Pinning
Content: View full analysis
Remediation: View remediation
\ --hash=sha256: ``` The same controls should be applied to `pandas` and all resolved transitive dependencies. ]]>

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 93% confidence
Finding: The skill clearly instructs file read/write operations over user working directories and spreadsheet files, but it does not declare any explicit tool scope or permission boundary. That increases the chance of overbroad file access or unintended modification if the agent runtime grants broader filesystem capabilities than the task actually needs.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The trigger conditions are broad enough to activate on ordinary requests about volunteer-hour summaries, data cleanup, or spreadsheet processing, which could cause the skill to engage unexpectedly on unrelated files. Because the skill performs multi-file scanning and consolidation, accidental invocation could expose or process data the user did not intend to include.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill is designed to collect and normalize personal data including names, student IDs, phone numbers, colleges, and volunteer records from multiple spreadsheets, yet it provides no privacy warning, consent check, retention limit, or handling guidance for sensitive personal information. In this context, silent bulk processing of student data materially raises privacy and compliance risk, especially when scanning all matching files in a directory.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This markdown file presents all headings, mappings, and matching rules exclusively in Chinese. Under the policy rule for language/locale, forcing a specific language without user opt-in or a documented justification can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The module docstring is written entirely in Chinese and describes the script as being used by an AI agent in a conversational workflow, which implies a Chinese-only interaction context. There is no indication that users may choose another language or that the locale restriction is explicitly documented as a justified regional constraint.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.