T01 · Skill Instruction Hijacking
Note
- Location
SKILL.md:36- Finding
Mandatory Promotional Content Injected into User-Facing Reports
- Content
View full analysis
- Remediation
View remediation
Security audit
📊 Crypto Arbitrage Monitor - 套利信号监控
Security checks for vulnerabilities and agentic risk
Overview
This is a coherent crypto trading-signal skill with no persistence or credential access, but it needs Review because it can produce actionable financial recommendations from broad triggers and contains a material annualized-return error.
Install only if you intentionally want a Chinese-language ETH crypto arbitrage signal helper. Verify live funding rates, timestamps, fees, slippage, and annualized-return math independently, and do not rely on the template's 40% example for trading decisions. Be aware it may activate on generic arbitrage or funding-rate questions.
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
Findings (2)
other
Warning
- Location
SKILL.md:45- Finding
Tenfold Inconsistency in Advertised Annualized Arbitrage Return
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The English trigger “arbitrage” is highly generic and can appear in many benign contexts, including conceptual questions, historical discussion, or comparisons across markets. Because this skill performs specific searches and outputs actionable buy/watch/risk signals, such a generic trigger raises the chance of unintended invocation and overreaching behavior in a high-risk financial domain.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The English trigger “arbitrage” is highly generic and can appear in many benign contexts, including conceptual questions, historical discussion, or comparisons across markets. Because this skill performs specific searches and outputs actionable buy/watch/risk signals, such a generic trigger raises the chance of unintended invocation and overreaching behavior in a high-risk financial domain.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The activation rules are underspecified because the skill claims to trigger on any listed phrase without context boundaries, exclusions, or confirmation requirements. In a trading signal skill, this ambiguity is more dangerous than usual: accidental activation could lead the agent to generate pseudo-authoritative investment signals and recommendations from ad hoc search results without the user explicitly asking for that workflow.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 86% confidence
- Finding
The natural-language interface, trigger phrases, and operating instructions are all presented in Chinese, which effectively imposes a language expectation without documenting user choice. Under the stated policy, forcing a specific language without opt-in can be a locale/language policy issue.
- Content
Static analysis
No suspicious patterns detected.