ClawHub
Back to skill

Security audit

Feishu Cli Chat

Security checks for vulnerabilities and agentic risk

Overview

This skill openly helps an agent read and manage Feishu chats, but it grants broad user-level access to private messages and group controls without enough built-in authorization checks.

Install only if you want the agent to operate with the logged-in Feishu user's chat access. Use the narrowest Feishu scopes possible, avoid using it for third-party private conversations unless clearly authorized, and manually review high-impact actions such as member changes, pinning, deletion, or group dissolution.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Ssd 3

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill explicitly instructs the agent to retrieve, inspect, and summarize chat histories, including content about who said what and thread replies, using a User Token that can access chats the bot is not in and p2p messages. This creates a direct natural-language exfiltration path where a user can ask the agent to reveal sensitive conversation content that the token holder can access, even if the requester should not see it.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

This section operationalizes how to recover private-chat records about another person through search, sender filtering, and message-detail lookup, despite direct API limitations. In context, that guidance materially lowers the barrier for surveillance or unauthorized review of direct-message content and makes the skill more dangerous than a generic chat viewer because it specifically enables targeted person-centric retrieval.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation criteria are broad enough to invoke the skill whenever a user mentions a chat name or expresses general interest in what a group discussed, even without a clear consent or purpose check. Because this skill exposes message history, private conversations, and chat-management actions via a User Token, over-broad routing increases the chance of unnecessary access to sensitive communications.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The documentation describes browsing chat history and managing conversations without prominently warning that the content may include sensitive personal, confidential, or regulated information. In an agent setting, omission of privacy safeguards can normalize casual retrieval and summarization of conversation data, making accidental disclosure to an unauthorized requester more likely.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.