Back to skill

Security audit

Kling AI MCP (Global)

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign verdict with high confidence: the findings are considered consistent with the skill’s stated use of Kling’s connected services, though they flag user content transmission and phrases about tokens and user confirmation.

The verdict is benign, but review the privacy implications before installing: prompts and media used with connected Kling tools may be sent to Kling. Avoid sending sensitive content unless you are comfortable sharing it with that service.

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
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 12)May include surrounding context.

md
`https://kling.ai/mcp/plugin`

Connect it through the host application's MCP settings and complete Kling's OAuth sign-in. Do not ask the user to paste passwords, access tokens, or cookies into chat. This skill does not install a local MCP server or CLI.

After connecting, call `tools/list` and `who_am_i` to discover the live tools, model names, accepted inputs, limits, and account region. Do not guess or reuse stale model settings. Use the MCP tools exposed by the connected server for image generation, video generation, uploads, Elements, motion-library lookup, account/credit queries, and task status.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 12)May include surrounding context.

md
`https://kling.ai/mcp/plugin`

Connect it through the host application's MCP settings and complete Kling's OAuth sign-in. Do not ask the user to paste passwords, access tokens, or cookies into chat. This skill does not install a local MCP server or CLI.

After connecting, call `tools/list` and `who_am_i` to discover the live tools, model names, accepted inputs, limits, and account region. Do not guess or reuse stale model settings. Use the MCP tools exposed by the connected server for image generation, video generation, uploads, Elements, motion-library lookup, account/credit queries, and task status.

The workflow instructs use of remote generation and upload tools without warning that user content is transmitted to Kling.

Medium
Category
Output Handling
Confidence
87% confidence
Finding

The skill instructs the agent to use the connected server for generation and uploads, which can transmit user prompts and media to a third-party service. It discloses credit usage but provides no corresponding privacy warning about sending that content to Kling.

Content

No source excerpt is available for this finding.