Einstein Research — Market Breadth Analyzer

• Outcome: Review
• Latest audit: Sep 14, 2026 · 11:22 PM UTC
• Version: 0.1.0

Overview

The skill appears to be a real market-breadth analyzer, but it needs review because it can fetch arbitrary URLs, write reports/history locally, and gives fairly prescriptive investment guidance.

Install only if you are comfortable with a financial-analysis skill that runs Python, uses outbound network access, and writes reports/history locally. Prefer running it with the default TraderMonty URLs in an isolated workspace, avoid passing untrusted custom URLs, and treat the equity exposure guidance as informational rather than investment advice.

A.I.G

Vulnerability Patterns

• Insecure Dependencies: Introduces malicious components through unsafe dependency sources
• 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

Findings (3)

T09 · Insecure Skill Coding Practices

Warning

• Location: scripts/market_breadth_analyzer.py:53

Finding

Arbitrary URL Fetching Enables Server-Side Request Forgery

Content

View full analysis

Remediation

View remediation

T09 · Insecure Skill Coding Practices

Warning

• Location: scripts/csv_client.py:50

Finding

Unbounded Remote CSV Downloads Can Exhaust Memory and CPU

Content

View full analysis

Remediation

View remediation

T08 · Insecure Dependencies

Note

• Location: README.md:20

Finding

Unpinned and Unnecessary Dependency Installation Creates Supply-Chain Risk

Content

View full analysis

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 92% confidence

Finding

A skill that claims to fetch TraderMonty data and compute a 0-100 breadth score but instead performs testing/report validation or otherwise lacks the primary analysis logic is a material integrity problem. In this context, the danger is less classic code execution and more deceptive or unreliable operation: users may receive financial-health conclusions that are ungrounded in the stated data source or methodology, while still granting network/file privileges based on the advertised purpose.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 96% confidence

Finding

A skill that claims to fetch TraderMonty data and compute a 0-100 breadth score but instead performs testing/report validation or otherwise lacks the primary analysis logic is a material integrity problem. In this context, the danger is less classic code execution and more deceptive or unreliable operation: users may receive financial-health conclusions that are ungrounded in the stated data source or methodology, while still granting network/file privileges based on the advertised purpose.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

A skill that claims to fetch TraderMonty data and compute a 0-100 breadth score but instead performs testing/report validation or otherwise lacks the primary analysis logic is a material integrity problem. In this context, the danger is less classic code execution and more deceptive or unreliable operation: users may receive financial-health conclusions that are ungrounded in the stated data source or methodology, while still granting network/file privileges based on the advertised purpose.

Content

No source excerpt is available for this finding.

Anti-Refusal Statement

High

• Category: Anti-Refusal
• Confidence: 80% confidence

Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · scripts/calculators/bearish_signal_calculator.py (reported line 15)May include surrounding context.

python

Scoring (100 = healthy):
  Base:
    Signal=False & Trend=1  -> 85 (all clear)
    Signal=False & Trend=-1 -> 50 (no warning but downtrend)
    Signal=True  & Trend=1  -> 30 (warning in uptrend)
    Signal=True  & Trend=-1 -> 10 (fully bearish)

Anti-Refusal Statement

High

• Category: Anti-Refusal
• Confidence: 80% confidence

Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · scripts/calculators/bearish_signal_calculator.py (reported line 108)May include surrounding context.

python

if not signal_active and trend == 1:
        return 85  # All clear
    elif not signal_active and trend == -1:
        return 50  # No warning but downtrend
    elif signal_active and trend == 1:
        return 30  # Warning in uptrend
    else:  # signal_active and trend == -1

Undeclared Tool Scope

Medium

• Category: MCP Least Privilege
• Confidence: 82% confidence

Finding

The skill declares no tool scope or permissions, yet its documented workflow explicitly invokes a Python script that downloads remote data and writes JSON/Markdown reports to disk. This creates an authority/visibility gap: hosts or reviewers may treat the skill as low-privilege while it actually depends on network and file-write capabilities, increasing the chance of unintended data access, persistence, or unsafe execution in a broader agent environment.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The module’s documented scoring rules diverge from the actual implemented logic, which can cause downstream users, reviewers, or other components to make decisions based on incorrect assumptions about how market breadth health is computed. In a financial-analysis skill, this is a real integrity issue because users may rely on the published scoring model to interpret signals, validate outputs, or compare results, and the mismatch can silently produce misleading assessments.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

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

Finding

The scorer returns explicit trading posture recommendations such as equity exposure ranges, profit-taking, hedging, and sector rotation without any built-in disclaimer, suitability check, or requirement for independent validation. In a skill marketed for assessing market health, this can cause users to treat the output as prescriptive investment advice and make harmful financial decisions based on incomplete or stale data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 80% confidence

Finding

The helper writes a temporary markdown file, reads it back, and unlinks it in cleanup. Although this is test code, the safety-relevant file operations have no visible user-facing notice beyond a generic docstring, so the behavior is not explicitly disclosed under the code-file warning criterion.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.