Back to skill

Security audit

Chanlun Analysis Pro

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: scanners found no reported malicious behavior, but flagged an unpinned installation command and potentially overbroad activation terms, with inspection coverage incomplete.

Before installing, consider whether you trust the source of the `clawhub` package and can use a reviewed exact version. The scanners also flagged activation terms that may be too broad. Treat the benign verdict as a reasoned assessment, not a guarantee: Skillspector inspected 80% of components, with a malformed manifest and unresolved references contributing to incomplete coverage.

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
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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
Findings (4)

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

The installation example invokes an unpinned npm package through npx, allowing the executable version to change between installations. Although this is a CLI rather than an MCP server, a compromised future release could execute code with the installing user's permissions; no compromise is demonstrated here.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx clawhub'.

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

The Chinese quick-start repeats the unpinned npx invocation, exposing users to executable package updates that have not necessarily been reviewed. The finding's MCP classification is inaccurate, but the underlying supply-chain risk remains: a malicious resolved release could run with the user's permissions.

Content

No source excerpt is available for this finding.

Trigger keywords include ambiguous everyday and cross-domain terms without requiring a Chanlun or stock-analysis context.

Medium
Category
Trigger Abuse
Confidence
98% confidence
Finding

The explicit trigger lists include ambiguous terms such as 'stroke', 'divergence', '笔', and '上涨' without requiring financial context. If used for automatic routing, these could activate trading-analysis instructions for unrelated requests and contaminate responses; the surrounding Chanlun description reduces this risk but does not explicitly prevent it.

Content

No source excerpt is available for this finding.

The drawing shortcut uses generic triggers without limiting them to Chanlun analysis.

Low
Category
Trigger Abuse
Confidence
90% confidence
Finding

The trigger column lists “画图/图示” (draw a diagram/illustration), which also occur in ordinary requests unrelated to financial analysis. Although the output is a Chanlun structure diagram, the activation condition does not explicitly require that context, so unrelated drawing requests could invoke this skill.

Content

No source excerpt is available for this finding.