同花顺问财数据提取器

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 3:59 AM UTC
• Version: 1.0.0

Overview

This is a disclosed scraper/exporter for iwencai stock data, but users should know it controls a debug Chrome session and creates local SQLite and Excel files.

Install only if you intend to automate a local Chrome debug session against iwencai and create local market-data files. Use a separate Chrome debug profile, review the configured DB_PATH and OUT_PATH before running, and treat generated spreadsheets as untrusted reports because scraped text could become spreadsheet formulas unless sanitized.

A.I.G

Vulnerability Patterns

• Insecure Skill Coding Practices: Finds exploitable flaws such as hardcoded secrets or command injection
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

• Location: scripts/export_excel.py:254

Finding

Excel Formula Injection Through Unsanitized Remote Data

Content

View full analysis

3 else '', 'price': clean_number(row[4]) if len(row) > 4 else None, 'change_pct': clean_number(row[5]) if len(row) > 5 else None, 'zt_time': row[6] if len(row) > 6 else '', 'zt_status': row[7] if len(row) > 7 else '', 'volume': parse_amount(row[9]) if len(row) > 9 else None, 'amount': parse_amount(row[10]) if len(row) > 10 else None, 'first_zt_time': row[11] if len(row) > 11 else '', 'lb_count': lb_count, 'zt_type': row[13] if len(row) > 13 else '', 'float_mv': parse_mv(row[14]) if len(row) > 14 else None, 'vol_ratio': clean_number(row[15]) if len(row) > 15 else None, 'themes': row[16] if len(row) > 16 else '', 'zt_tags': row[17] if len(row) > 17 else '', 'total_mv': parse_mv(row[19]) if len(row) > 19 else None, }) ``` The exporter writes those values directly into workbook cells: ```python for i, row in df[detail_cols].iterrows(): r = 3+i; bg = RED_LIGHT if i%2==0 else WHITE for j, col_name in enumerate(detail_cols,1): c = ws4.cell(row=r, column=j); c.value = row[col_name] dat(c, ...[truncated 3264 chars]

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (10)

Undeclared Tool Scope

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence

Finding

The skill clearly instructs the agent to run local commands (cmd /c python ...) and invoke agent-browser, but the manifest does not declare any explicit tool scope or allowed-tools restriction. That creates an authorization gap where a broad shell-capable agent may execute filesystem and browser-automation actions without the skill formally constraining or disclosing that behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 91% confidence

Finding

The manifest uses broad trigger terms like 问财, 涨停数据, 爬取涨停, and AI选股数据, which can match ordinary finance questions rather than an explicit request to automate browsing and persist scraped data locally. This increases the chance the skill is auto-invoked in contexts where the user did not intend browser control, scraping, or local data collection.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 89% confidence

Finding

The skill persistently writes scraped data to a local SQLite database and exports Excel reports to local paths, but the description does not prominently warn that it creates and updates local files. Without an explicit warning, users may not understand that invoking the skill causes durable data storage and report generation on the host system.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 85% confidence

Finding

The helper function invokes an external command via subprocess to control a browser session, and later actions use it to click, fill, press keys, and evaluate JavaScript on the user's open browser page. Although the module docstring states prerequisites, it does not clearly warn the user that the script will actively automate and manipulate the browser session.

Content

No source excerpt is available for this finding.

subprocess module call

Medium

• Category: Dangerous Code Execution
• Confidence: 70% confidence

Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/crawl.py (reported line 63)May include surrounding context.

python

if not args:
        return ""
    cmd = [AGENT, "--cdp", CDP_PORT] + list(args)
    r = subprocess.run(cmd, capture_output=True)
    out = r.stdout.decode("utf-8", errors="replace").strip()
    if not out and r.stderr:
        err = re.sub(r'\x1b\[[0-9;]*m', '', r.stderr.decode("utf-8", errors="replace").strip())

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

Natural-language strings throughout the script, including the module description and generated workbook labels, are fixed in Chinese with no opt-in or locale selection. This can violate language or locale policy when users are not given a choice or the locale restriction is not explicitly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The entire troubleshooting guide is written in Chinese, including headings and operational instructions, with no indication that users may choose another language or that the skill is intended only for a Chinese-speaking or region-specific audience. Under the policy for natural-language violations, forcing a specific language without opt-in is in scope.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

• Category: Not specified by scanner
• Confidence: 90% confidence

Finding

The module docstring presents the skill description and usage entirely in Chinese and assumes a Chinese-language target workflow, with no indication that users may choose another language or locale. Under the stated policy, forcing a specific language without opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 81% confidence

Finding

The script opens the configured SQLite database and creates/inserts records, which modifies a local file on disk. While the database path is configurable, there is no explicit user-facing warning in the script that running it will create or alter the database contents.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 86% confidence

Finding

This code file performs a file write by saving a workbook to the path in OUT_PATH. Although the module docstring mentions the output path, the runtime action itself has no confirmation prompt or explicit warning before overwriting or creating the file.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.