Security audit

feishu-doc-sender

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a document-sending helper, but its sending behavior and authority are not clearly implemented or scoped.

Review this skill before installing if you expect real Feishu/Lark delivery. Confirm exactly which tool or API will upload files, which workspace paths it can scan, and require explicit confirmation of both the selected file and recipient before any document is sent.

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

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (9)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 98% confidence
Finding: The skill claims to send documents to Feishu/Lark users and groups, but the described behavior does not show any actual Feishu API interaction and appears limited to local file discovery/listing. This mismatch can mislead users and downstream agents into trusting the skill with document delivery workflows, causing accidental disclosure, failed delivery assumptions, or unsafe chaining with other tools to compensate for missing functionality.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The manifest description limits the skill to sending Word (.docx) and PDF documents, presenting that as the skill's scope. However, the same file documents later describe EPUB detection/support as part of file filtering and supported formats, which expands behavior beyond the manifest's stated scope.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This skill file includes natural-language instructions in Chinese, and the usage/examples throughout the document are also Chinese-centric. Because the file does not offer a user language choice or explain that the skill is intentionally region/language-specific, it may violate language/locale policy expectations.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger phrase "send document" is broad enough to match ordinary user requests that may not specifically intend to invoke this skill. In a messaging/file-transfer skill, accidental invocation can cause unintended discovery and sending of local .docx/.pdf files to Feishu recipients, increasing the risk of data leakage or surprising side effects.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The trigger phrase "send word" is ambiguous and can overlap with common conversational language, making unintentional skill activation plausible. Because this skill automatically detects document files in the workspace, mistaken activation could expose sensitive documents or initiate unwanted outbound file transfers.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The README presents the skill description, usage examples, features, and troubleshooting entirely in Chinese, which effectively imposes a specific language on users. Under the policy, language constraints should either be optional or clearly justified as region-specific.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 76% confidence
Finding: The documentation describes '一键发送' (one-click send), implying immediate delivery, while later sections say the skill lists found files and asks for confirmation before sending. This is an intent-level contradiction in the skill's own documentation about whether sending is automatic or confirmation-gated.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The script emits user-facing status text in Chinese strings only (for both the no-results and results cases). This imposes a specific language on users without opt-in or an alternative locale selection, which matches the language/locale policy concern.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.