Back to skill

Security audit

deepseek

Security checks for vulnerabilities and agentic risk

Overview

The skill clearly describes sending DeepSeek requests through RunAPI, with bounded retries and no hidden installation or background activity.

Prompts, messages, and supported images are sent to RunAPI for processing. Confirm that RunAPI is approved for your data, use a RunAPI-specific key, and scope environment settings to the intended client so other applications are not redirected. API usage may incur charges.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

The request workflow sends prompt content to RunAPI without a privacy warning.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The skill instructs the agent to send model requests to RunAPI, which transmits the supplied prompt content to a third-party service. Although the provider and endpoint are identified, the skill does not warn that user content leaves the local environment or caution against submitting sensitive data.

Content

No source excerpt is available for this finding.

The compatibility instructions redirect existing clients to RunAPI without warning that request content will be sent to a third-party provider.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

These instructions configure existing Anthropic Messages and Gemini clients to use RunAPI endpoints with message content and potentially images. Although the destination URLs are shown, the reference does not warn that user content will be processed by RunAPI rather than the original provider, which can affect privacy and data-handling expectations.

Content

No source excerpt is available for this finding.