Security audit

Feishu File Send

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Feishu file-sending helper, with no hidden persistence or bundled executable payloads, though users should be careful not to send sensitive files unintentionally.

Install only if you intend agents to send files through Feishu. Confirm the exact file path and contents before use, especially for workspace files that may contain private, personal, or business-sensitive 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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The documentation encourages sending arbitrary files to Feishu but does not clearly warn that file contents leave the local environment and may expose secrets, personal data, or internal documents. In an agent setting, this omission can normalize exfiltration of sensitive workspace files under the guise of routine messaging.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: The skill is documented as a file-sending helper, but it also includes a subprocess-based workflow that generates a PDF by executing Python code. This expands the skill from simple transmission into code execution guidance, increasing attack surface and making it easier for an agent to perform unintended local actions before exfiltrating the result.

Natural-Language Policy Violations

Low

Confidence: 88% confidence
Finding: SQP-3 applies to all file types and includes language or locale policy violations. The operational description and instructions are presented in Chinese only, with no indication that users may choose another language or that the skill is restricted to a Chinese-language context.

Static analysis

No suspicious patterns detected.