ClawHub
Back to skill

Security audit

Llm Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is an instructional guide for building Runware-hosted LLM agents, and its external API use is disclosed and aligned with that purpose.

Before installing, understand that using this skill means prompts, message history, tool schemas, and tool results may be sent to Runware's hosted API with your Runware API key. Avoid sending sensitive data unless that third-party processing is acceptable for your use case, and keep any exposed tools narrowly scoped.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

md
# LLM agent worked recipes

Three end-to-end recipes for the `llm-agent` skill. Each shows the scenario, the request, and the response in **OpenAI chat-completions format** over Runware's OpenAI-compatible endpoint at `https://api.runware.ai/v1`. This is not the `imageInference` task shape. Point any OpenAI client at the base URL with your Runware API key and set `model` to the AIR.

Field reminders:

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 3)May include surrounding context.

md
# LLM agent worked recipes

Three end-to-end recipes for the `llm-agent` skill. Each shows the scenario, the request, and the response in **OpenAI chat-completions format** over Runware's OpenAI-compatible endpoint at `https://api.runware.ai/v1`. This is not the `imageInference` task shape. Point any OpenAI client at the base URL with your Runware API key and set `model` to the AIR.

Field reminders:

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

This markdown file documents use of Runware's remote OpenAI-compatible endpoint and instructs readers to point a client at that URL with an API key. Because the examples involve transmitting prompts and tool context to an external service, the description should explicitly warn that request contents and metadata are sent over the network to a third-party endpoint.

Content

No source excerpt is available for this finding.