Security audit

Feishu Workspace

Security checks for vulnerabilities and agentic risk

Overview

This is a clear Feishu workspace helper, with expected read/write workspace behavior and no hidden code, install hooks, or persistence.

Install only with Feishu permissions appropriate to the work you want automated. Before letting the agent write to shared docs or Bitable tables, verify the target resource and review generated content, especially when chat content contains confidential, unverified, or sensitive team information.

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
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 (4)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The README repeatedly encourages writing meeting notes, summaries, action items, and project updates directly into Feishu based on chat content, but it does not warn users that chat data may be inaccurate, sensitive, or unauthorized for publication. In this skill context, the omission is meaningful because the core function is transforming conversational input into persistent documents and trackers, which can lead to data leakage, incorrect records, or unintended modification of shared workspace content.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill explicitly authorizes writing or updating Feishu docs and Bitable records but provides no instruction to confirm user intent before modifying external workspace data. In an agent setting, that omission can lead to unintended state-changing actions, including overwriting shared documents or creating inaccurate records in collaborative systems.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The example prompts repeatedly instruct the agent to write, append, or create content in Feishu shared resources without any warning, confirmation step, or reminder that these actions modify collaborative documents and records. In a workspace integration, this can normalize unsafe behavior and increase the chance of unintended edits, data corruption, or accidental disclosure in shared team assets when users invoke examples verbatim.

Missing User Warnings

Low

Confidence: 87% confidence
Finding: The skill encourages searching internal wiki knowledge before answering but does not warn that this may access sensitive internal workspace content. Without guidance on consent, scope limitation, or summarization boundaries, an agent could retrieve or expose internal information more broadly than the user intended.

Static analysis

No suspicious patterns detected.