Feishu Literature Manager

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:28 PM UTC
• Version:
  1.1.1

Overview

This skill does what it says: it searches PubMed and creates or updates Feishu literature tables, with the main caution that it can write records and grant the requester full table access.

Install only if you want an agent to create or update Feishu Bitable literature databases. Before running it, confirm the target Feishu table or folder, search topic, number of papers, and whether full_access should be granted; use it only with Feishu tables you control.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (3)

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill is designed to create fields, add records, and modify Feishu Bitable content, but the upfront description does not clearly warn users that it will write to and alter their Feishu data. In a skill that can create or update user-owned tables, lack of explicit write-disclosure can lead to surprising or unauthorized modifications.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The workflow instructs automatically setting table permissions to full_access without an explicit consent step. Permission changes are security-sensitive because they alter who can manage, share, and delete user data; doing this by default expands access and control beyond the minimum needed.

Missing User Warnings

High

• Confidence: 98% confidence
• Finding: This section mandates always granting full_access and even documents enabling the permission tool, normalizing a privileged action without user opt-in or contextual necessity. In skill context, this is more dangerous because the same workflow also creates and populates user data stores, so automatic privilege changes can compound the impact of mistakes or misuse.

VirusTotal

64/64 vendors flagged this skill as clean.