Security audit

Ta Analyzer

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed crypto technical-analysis skill that fetches Binance market data and returns trading-style signals, with financial-risk caveats users should consider but no evidence of hidden or malicious behavior.

Install only if you want an agent to fetch public Binance market data and produce technical-analysis-based trading signals. Treat its Buy/Sell, stop-loss, take-profit, and price-prediction outputs as informational and uncertain, not professional financial advice, and be aware that some explanations may appear in Chinese.

A.I.G

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

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 96% confidence
Finding: The skill metadata and documentation frame the capability primarily as technical analysis, but the file also explicitly promises BUY/SELL recommendations, stop-loss, take-profit targets, and price-prediction style outputs. That mismatch can cause an orchestrator or user to invoke the skill under a lower-risk assumption, while it actually performs higher-risk financial-advisory behavior; hardcoding Binance also narrows data provenance and may produce exchange-specific outputs without clear disclosure.
Content: No source excerpt is available for this finding.

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 72)May include surrounding context.
md
const { analyze } = require('./index.js');

Ae1

High

Category: analysis-evasion
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected
Content: Scanner excerpt · SKILL.md (reported line 78)May include surrounding context.
md
const { analyze } = require('./index.js');

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The activation text is broad enough to match common requests such as chart analysis, indicator values, trading signals, or price predictions, increasing the chance the skill is auto-invoked in contexts where the user did not clearly request financial guidance. In a trading-related skill, overbroad routing is riskier because it can expose users to unsolicited or overconfident financial recommendations.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The documentation advertises trading recommendations and price predictions without any caution about financial risk, uncertainty, or non-advisory limitations. In this skill context, that omission is dangerous because users may treat the outputs as reliable investment advice and act on them without understanding model, market, or exchange-specific limitations.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The code emits analysis strings in Chinese for volume analysis regardless of user preference or locale. This is a natural-language policy issue because it forces a specific language without any opt-in or documented region-specific justification.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The manifest describes a technical analysis skill for indicators, chart analysis, trading signals, and price predictions. In addition to analysis outputs, the code synthesizes an actionable recommendation object with BUY/SELL/NEUTRAL decisions plus stop-loss and take-profit levels, which is a stronger portfolio/trade-advice behavior than the analysis-oriented description suggests.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The recommendation summary and reasons use Chinese phrases such as trend, pattern, and waiting-for-signal messages with no language negotiation. This forces a specific language in user-facing output and violates the language/locale policy criteria.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 81% confidence
Finding: The header comment presents the skill as broadly supporting technical analysis, and the manifest specifically claims detection of 10+ chart patterns. The code implements hammer, shooting star, engulfing, doji, morning star, double top, double bottom, head and shoulders, inverse head and shoulders, triangle, and flag; depending on whether triangle subtypes are counted separately, this may only just satisfy or ambiguously undershoot the claim.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.