ClawHub
Back to skill

Security audit

博客发布规则 本项目是 astro-koharu 博客。博客发布规则

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly scoped blog-publishing helper that writes new Astro Markdown posts and runs the site build, with limits against overwrites, deletion, secrets access, and unrelated commands.

Install this only if you want the agent to create new posts in the named Astro blog directory and run the site build. Review publish requests before invoking it, because a successful build can update what the site serves.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description contains broad trigger phrases like 'publish', 'generate', 'classify', and generic Chinese blogging requests, which can cause the skill to activate for routine content tasks without clearly signaling that it will write to the server filesystem and run a build. In this context, over-broad activation is risky because the skill has side effects on a live site directory, so accidental invocation could modify content or trigger builds unexpectedly.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The workflow instructs the agent to create files under a production web root and run pnpm build, but the skill does not prominently warn users that invoking it will perform server-side writes and execute a build command. This raises the chance of unintended destructive or operational side effects because users may think they are only asking for content generation, while the skill is authorized to change deployed content.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Line L140 requires generated descriptions to be in Chinese, which imposes a specific language policy on outputs. The file does not offer the user a language choice or explain a necessary region-specific reason for this restriction.

Content

No source excerpt is available for this finding.