Back to skill

Security audit

Knowledge Orchestrator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed knowledge-search compatibility wrapper for Zotero, Obsidian, and IMA, with no packaged evidence of hidden execution, persistence, destructive behavior, or exfiltration.

Install only if you intend to connect local knowledge sources such as Zotero and Obsidian. Keep API keys out of shared repositories, restrict access to credential files, and review notes before using any future IMA sync/upload feature because private research content may be sent to external services.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · README.md (reported line 1)May include surrounding context.

md
# Knowledge Orchestrator 鎶€鑳?
**鐗堟湰锛?* v0.1  
**鍒涘缓鏃堕棿锛?* 2026 骞?3 鏈?26 鏃? 
**鐘舵€侊細** 馃攧 寮€鍙戜腑

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The README instructs users to create local files containing API credentials and describes syncing/uploading notes to an external cloud service, but it does not warn about secure storage, accidental disclosure, or the sensitivity of transmitted note content. In an agent skill context, this can lead users to expose tokens in predictable paths and send research notes or knowledge-base data to third-party services without informed consent or data-handling safeguards.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file presents the description, warnings, and examples entirely in Chinese, including the required argument format and operational guidance, with no indication that other languages are supported. This can constitute a language/locale policy violation when no user opt-in or language choice is offered.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The manifest's argument hint is simply "[操作] [参数]", which is extremely broad and does not constrain what user phrasing should invoke the skill. In a markdown/manifest context, this lacks specificity about supported commands or exclusions and could overlap with many routine requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This code file contains natural-language descriptions and CLI help text entirely in Chinese, including the module docstring and usage examples. Under the policy, forcing a specific language without user opt-in is a violation unless the locale constraint is explicitly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The description is written only in Chinese ("知识编排与管理技能"), which may impose a language choice without offering any user opt-in or alternative locale. The policy requires avoiding forced language or locale constraints unless explicitly documented and justified.

Content

No source excerpt is available for this finding.