Security audit

Cup and Handle Pattern Detector

Security checks for vulnerabilities and agentic risk

Overview

This stock-pattern skill behaves like its description: it analyzes market data and saves local charts or reports, with one contained SQLite input-validation issue users should understand.

Install only if you are comfortable providing a Tushare token for market-data retrieval and writing generated charts/reports to local temp-style directories. For batch scans, use databases you trust or that follow the expected stock-table schema because crafted table names could affect query behavior or results.

A.I.G

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)

T09 · Insecure Skill Coding Practices

Warning

Location: scripts/batch_scan_cup_handle.py:65
Finding: SQL Injection Through Unquoted SQLite Table Identifier
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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 (12)

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.
md
python3 scripts/cup_handle_detector.py <股票代码>

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 54)May include surrounding context.
md
python3 scripts/cup_handle_detector.py <股票代码>

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 48)May include surrounding context.
md
python3 scripts/batch_scan_cup_handle.py --db <数据库路径>

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.
md
python3 scripts/batch_scan_cup_handle.py --db <数据库路径>

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill advertises executable workflows that read environment variables (TUSHARE_TOKEN) and write generated chart files to a temporary directory, but the manifest does not declare any explicit tool scope or permissions. This creates an authorization and transparency gap: a caller or platform reviewer cannot tell from the skill declaration that it may access secrets or write files, which can lead to over-privileged execution or unexpected data handling.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The manifest description is entirely in Chinese, while the skill title and trigger set include English terms, suggesting mixed-language use but no explicit user choice of language. The policy requires avoiding language forcing unless the user is offered a language/locale choice or the constraint is clearly justified.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The script enumerates every table in the SQLite database and scans/ranks all stocks, which materially exceeds the skill description of answering stock-specific cup-and-handle queries. In an agent environment, this scope expansion can cause unintended bulk data processing, higher resource consumption, and broader-than-expected access to locally available market data.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This Python skill is a code file, so SQP-3 applies. The module docstring and subsequent user-facing strings are primarily Chinese by default, and the code automatically uses Chinese labels when a Chinese font is available rather than offering the user a language choice; this can violate language/locale policy when no opt-in is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file uses a single fixed language throughout, and there is no indication that users may opt into Chinese or that the skill is region-specific. Under the policy rule, forcing a specific language without user choice is a natural-language policy concern.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Low

Category: Not specified by scanner
Confidence: 80% confidence
Finding: The script persists a markdown report summarizing scan results in addition to generating charts, which is broader than the stated illustration-focused behavior. In a shared agent workspace, unexpected persistent outputs can create data retention and information exposure risks, especially if report contents are later consumed by other tools or users.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Low

Category: Not specified by scanner
Confidence: 73% confidence
Finding: The manifest says the skill detects the pattern and generates an illustrative image, but the implementation persists that image under ~/temp by default and creates directories as needed. File creation may be an implementation detail, but the persistent write behavior is not explicit in the declared description.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Low

Category: Not specified by scanner
Confidence: 78% confidence
Finding: The manifest describes a cup-and-handle detector for daily stock charts, but does not mention dependency on external credentials or runtime access to environment-stored API tokens. Reading TUSHARE_TOKEN and initializing a remote API client adds a credential-handling capability beyond the stated analytical purpose.
Content: No source excerpt is available for this finding.