Back to skill

Security audit

conversation-exporter

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local export tool, but it broadly reads persistent WorkBuddy memory and operational metadata and writes history files without clear source-by-source consent or concrete redaction rules.

Install only if you are comfortable with exports that may include persistent WorkBuddy memory, project metadata, automation-task information, and Skill usage records. Before using it, specify the exact date range, sources, output format, and destination, and review the generated files for secrets or private information before sharing them.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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
Findings (1)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:45
Finding

Overbroad Access to Persistent Memory and Operational Metadata

Content
View full analysis
Remediation
View remediation
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 (4)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

文件描述和整份技能说明均以中文固定呈现,没有说明可根据用户语言偏好切换,也没有注明该技能仅面向特定中文环境。这构成了语言/locale 约束未获用户选择或明确合理化的自然语言策略问题。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The trigger conditions are broad enough to match common requests such as generating summaries, analyzing conversation data, or archiving memory, which may invoke the skill when the user did not intend a persistent export. In this skill's context, unexpected activation is more dangerous because it reads memory files and writes history artifacts to the workspace, increasing the chance of privacy leakage or unintended data retention.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill claims it will automatically filter sensitive information and only export summaries, but the documented export content includes a full chronological timeline of interactions, work memory, project information, milestones, and skill usage records. This mismatch can cause operators or users to overtrust the privacy properties of the export and unintentionally write sensitive historical data to disk.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
72% confidence
Finding

L050-L055 把“导出前验证数据完整性”列为安全特性,但在使用流程、错误处理和技术实现部分(如 L074-L084、L205-L223)只描述了读取、构建和写入,没有对应的验证逻辑或检查标准。虽然这更像缺失实现说明,但在当前文件中属于对能力的主动宣称与实际描述步骤不一致。

Content

No source excerpt is available for this finding.