Back to skill

Security audit

Stocks and Financial Data Pull Crypto Finance Cryptocurrency Price Earnings Dividends

Security checks for vulnerabilities and agentic risk

Overview

The skill’s access fits its market-data purpose, though confirmed reporting bugs require users to check its financial output carefully.

Searches and ticker requests go to Yahoo Finance. Check figures and reporting periods independently: get_income_statement with period=ttm can return annual figures, and get_complete_analysis omits most supporting results. These verified reliability defects do not establish a security hold. Use the documented pinned setup; review and pin a version before choosing the alternative uvx installation.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (10)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Dynamic import via __import__()

Medium
Category
Dangerous Code Execution
Confidence
75% confidence
Finding

Dynamic import() can load arbitrary modules at runtime, bypassing static analysis and potentially importing malicious code.

Content

Scanner excerpt · scripts/cli.py (reported line 60)May include surrounding context.

python
logging.basicConfig(level=logging.WARNING)
    sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
    try:
        module = __import__(MODULE)
    except ImportError as e:
        print(f"Missing dependency ({e}). Set up the skill first:\n  {SETUP}", file=sys.stderr)
        return 2

MCP server referenced without pinned version: 'uvx yfinance-ai'.

Medium
Category
MCP Rug Pull
Confidence
92% confidence
Finding

The documented command uses yfinance-ai@latest, a mutable release selector rather than a pinned version. If followed, it can execute a future unreviewed or compromised package with the launching user's permissions; however, this file only documents the command and does not execute it automatically.

Content

No source excerpt is available for this finding.

The income-statement tool labels annual financial statements as trailing-twelve-month data.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

For period='ttm', the tool fetches stock.income_stmt, which is the annual statement interface, and falls back to another annual interface while still labeling the result TTM. This creates misleading financial output that can distort valuation and investment analysis, although it is a data-integrity defect rather than an attacker-controlled code-execution path.

Content

No source excerpt is available for this finding.

The complete-analysis tool promises comprehensive results but unconditionally suppresses its main data sections.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

Every string contains the empty string, so the condition requiring "" not in response is always false and suppresses all main single-ticker results. Similar conditions suppress fund and market results, allowing a report presented as complete analysis to omit its supporting data, which is a meaningful integrity problem in a financial assistant rather than evidence of malicious execution.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/yfinance_ai.py (reported line 5380)May include surrounding context.

python
"""Fetch a Yahoo calendar DataFrame for the next N days"""
        days = max(1, min(int(days), 90))
        start = datetime.now().date()
        return getattr(yf.Calendars(), method)(start=start.isoformat(),
                                               end=(start + timedelta(days=days)).isoformat(),
                                               limit=max(1, min(int(limit), 100)))

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/yfinance_ai.py (reported line 5940)May include surrounding context.

python
for tool_name, args in tools:
                try:
                    method = getattr(self, tool_name)
                    response = await method(*args)

                    # Analyze response

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/yfinance_ai.py (reported line 6151)May include surrounding context.

python
}
                    })

                method = getattr(self, func_name)
                response = await method(ticker, *args)
                functions_run += 1

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/yfinance_ai.py (reported line 6174)May include surrounding context.

python
for func_name, args, label in fund_functions:
                try:
                    method = getattr(self, func_name)
                    response = await method(ticker, *args) if args else await method(ticker)
                    functions_run += 1

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/yfinance_ai.py (reported line 6248)May include surrounding context.

python
for func_name, args, label in market_functions:
            try:
                method = getattr(self, func_name)
                response = await method()
                functions_run += 1