ClawHub
Back to skill

Security audit

Chanlun Analysis Pro

Security checks for vulnerabilities and agentic risk

Overview

The skill is markdown-only and shows no malware-like behavior, but it should be reviewed because it gives concrete stock buy/sell and stop-loss guidance despite educational disclaimers.

Review before installing if you do not want an agent to produce specific trading entries, exits, stop-losses, or target prices. Treat all market examples as non-personalized and stale unless independently verified with your own data source, and avoid letting the skill execute or place trades without separate user confirmation and appropriate financial controls.

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 (5)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill opens with a disclaimer that it is not investment advice, but later provides concrete trading actions such as entry zones, stop-loss rules, high/low sell guidance, and positioning language. This inconsistency can mislead downstream agents or users into treating the content as safer or more educational than it really is, increasing the chance that actionable financial advice is delivered without appropriate controls.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The text says dates, prices, and securities are illustrative, but later sections present current-looking recommendations, expected buy zones, and tactical actions tied to specific instruments. That contradiction can cause users or calling systems to rely on supposedly 'example' numbers as live recommendations, defeating the safety disclaimer.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list is very broad and includes common terms like technical analysis, stock analysis, MACD, divergence, and Python Chanlun. This increases the chance of accidental invocation in general financial or programming conversations, causing unsolicited financial-analysis behavior in contexts where the user did not clearly request this skill.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Appendix guidance escalates from educational description into explicit trading directives such as 'aggressive buy,' 'escape top,' and concrete confirmation conditions. This makes the earlier educational-only framing unreliable and increases the likelihood that an agent will output direct trading instructions under the guise of analysis.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The phrase "中文触发词（优先）" indicates a built-in preference for one language in activation behavior. There is no accompanying statement that the user may choose language freely or that the preference is optional, so this can be read as forcing or biasing a specific language without opt-in.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.