Back to skill

Security audit

爆款决-自媒体操盘方法论

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with high confidence: the completed scan found no AIG issues, and the reported findings describe an unpinned install command and content-advice risks rather than malicious behavior.

Consider whether you are comfortable installing through an unpinned CLI, and review the installation changes before proceeding. Be cautious with any advice to delete posts because content and history may be unrecoverable. Skillspector reports complete coverage of all 12 components; AIG completed with no findings. A clean AIG result does not remove the risks noted above.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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
Findings (3)

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

Medium
Category
MCP Rug Pull
Confidence
86% confidence
Finding

The installation command invokes an unpinned npm CLI, allowing the package version executed during installation to change without review. A compromised future release could execute code with the installer's permissions; this is a conditional supply-chain exposure, not evidence of malicious code, and the finding incorrectly describes the CLI as an MCP server.

Content

No source excerpt is available for this finding.

The account-troubleshooting workflow recommends deleting old posts without warning about potential permanent data loss.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The account troubleshooting advice suggests deleting old posts to reset account labels without warning that posts, engagement, and associated history may be unrecoverable. Although it does not explicitly authorize autonomous deletion, its placement in an actionable diagnostic workflow can lead to destructive recommendations or actions based on an unverified explanation for poor performance.

Content

No source excerpt is available for this finding.

The trigger “我在坚持/我在积累” (“I’m persisting / I’m accumulating”) matches everyday statements without requiring a relevant content-creation or compounding-analysis request.

Low
Category
Trigger Abuse
Confidence
91% confidence
Finding

The instruction automatically routes these broad statements into a compounding analysis, even though users may be discussing unrelated habits, experiences, or goals. It provides no explicit contextual restriction or exclusions to distinguish relevant requests from incidental mentions.

Content

No source excerpt is available for this finding.