Back to skill

Security audit

World Cup Assist Value Trading

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent and disclosed, but it includes real-money trading capability driven by fragile web-search data and trading credentials without strong live-run controls.

Review before installing. Keep DRY_RUN enabled unless you have audited the strategy, use dedicated low-limit Simmer and Tavily keys, run it in an isolated environment, pin dependencies, and avoid scheduling live mode until cumulative spend and position limits plus explicit confirmation are added.

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)

T08 · Insecure Dependencies

Warning
Location
clawhub.json:4
Finding

Unpinned Trading SDK Creates a Supply-Chain Risk

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/signal.py:24
Finding

External Search Requests Lack Timeout, Error Handling, and Response Validation

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (15)

Tainted flow: 'TAVILY_API_KEY' from os.environ.get (line 6, credential/environment) → requests.post (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/signal.py (reported line 25)May include surrounding context.

python
return players

def tavily_search(query):
    r = requests.post("https://api.tavily.com/search", json={"api_key": TAVILY_API_KEY, "query": query, "max_results": 3, "include_domains": ["fbref.com", "espn.com", "whoscored.com"]})
    results = r.json().get("results", [])
    return " ".join([x.get("content", "") for x in results])

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill description says it 'prints buy signals', but the code contains a live trading path that executes client.trade(...) whenever DRY_RUN is changed. This creates a dangerous capability mismatch: a reviewer or operator may approve or run the skill believing it is signal-only, while it can place real trades with authenticated account access.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill documentation indicates it uses environment variables and network access, but it does not declare an explicit tool scope such as permissions or allowed-tools. That creates unnecessary ambiguity about what the agent is permitted to access, increasing the risk of over-privileged execution, unintended secret exposure, or broader network activity than reviewers expect.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

A general web search API provides broader external access than is necessary for a narrowly scoped task of reading football assist statistics for valuation. Even without direct evidence of abuse in this file, this overbroad capability can enable unintended data retrieval paths, reduce auditability, and violate least-privilege expectations for the skill.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest requests a Tavily API key and explicitly states it may fetch stats from FBref, ESPN, and WhoScored, while the skill description says the valuation is FBref-based. This mismatch broadens the data-access scope beyond what the user would reasonably expect, increasing the risk of unnecessary third-party data access and making the skill's behavior less transparent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The notes discuss future live trading and API execution without any accompanying warning about financial risk, real-money loss, or the need for explicit user authorization. In a financial/trading context, normalizing automated execution without risk disclosures can lead users to deploy unsafe automation and suffer monetary harm, especially given the notes also acknowledge unreliable data quality and placeholder calibration.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The heading labels 'Portugal (Lamine Yamal)', but the player annotation does not match the national-team context presented by the heading. This is an active documentation contradiction within the file, which can mislead downstream users or agents about which player/team the schedule refers to.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The script initializes an authenticated SimmerClient from environment credentials even though the stated purpose is only to print signals. Loading trading-capable credentials unnecessarily increases blast radius: any later code path, accidental modification, or imported behavior can use those credentials to act on the account.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/signal.py (reported line 25)May include surrounding context.

python
return players

def tavily_search(query):
    r = requests.post("https://api.tavily.com/search", json={"api_key": TAVILY_API_KEY, "query": query, "max_results": 3, "include_domains": ["fbref.com", "espn.com", "whoscored.com"]})
    results = r.json().get("results", [])
    return " ".join([x.get("content", "") for x in results])

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/signal.py (reported line 25)May include surrounding context.

python
return players

def tavily_search(query):
    r = requests.post("https://api.tavily.com/search", json={"api_key": TAVILY_API_KEY, "query": query, "max_results": 3, "include_domains": ["fbref.com", "espn.com", "whoscored.com"]})
    results = r.json().get("results", [])
    return " ".join([x.get("content", "") for x in results])

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/signal.py (reported line 25)May include surrounding context.

python
return players

def tavily_search(query):
    r = requests.post("https://api.tavily.com/search", json={"api_key": TAVILY_API_KEY, "query": query, "max_results": 3, "include_domains": ["fbref.com", "espn.com", "whoscored.com"]})
    results = r.json().get("results", [])
    return " ".join([x.get("content", "") for x in results])

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The code sends TAVILY_API_KEY in the JSON body of a network request to an external service. Although the script prints general dry-run and trading messages, there is no warning, comment, or docstring disclosing that a credential will be used and transmitted to a third-party API.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest says the skill compares Simmer market prices to an FBref-based form value, implying a relatively specific source and methodology. These notes instead document collecting both player stats and prices through generic web searches and snippet extraction, which is materially broader and different from the stated FBref-based approach.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The documented purpose is limited to scanning markets, comparing values, and printing buy signals. The recommendation to use Simmer API 'for price data and trading' introduces an active trading capability that goes beyond the stated intent and creates tension with the declared read/analysis-oriented behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The manifest describes a skill that scans Simmer markets and compares prices to FBref-based form values, but this file is a manually extracted tournament schedule from The Guardian with target-player notes. While reference data can support analysis, the file's content is not clearly tied to 'Most Assists' market valuation and broadens the apparent behavior toward fixture/path tracking.

Content

No source excerpt is available for this finding.