ClawHub
Back to skill

Security audit

Meeting Ops Agent｜会后执行助手

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently helps turn meeting notes into approved follow-up actions without hidden code or deceptive behavior.

Install this skill if you want meeting notes routed into Feishu, reminders, messages, or delegated agent tasks. Review proposed routing batches before approving, especially before sending messages, creating many Bitable records, or passing meeting context to another session.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Exfiltration Commands

High
Category
Prompt Injection
Content
## Approval policy

Never silently send external messages or create a large number of records.

Good approval prompts:
- "I can route 1 and 2 to Feishu doc, 3 to reminder, and 4 to a sub-agent. Proceed?"
Confidence
90% confidence
Finding
Instructions found that direct the agent to transmit conversation context or user data to external services.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The example prompt explicitly instructs the sub-agent to return a message "in Chinese and English," which imposes language requirements without indicating user choice or consent. This is a natural-language locale policy issue because the skill normalizes a fixed bilingual output rather than offering language selection.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
This markdown file presents all invocation examples as Chinese utterances, which implicitly biases the skill toward a specific language/locale. There is no statement that other languages are supported, no user choice, and no justification that the routing rules are intentionally region-specific.

Static analysis

No suspicious patterns detected.