Back to skill

Security audit

Wiro AI for n8n

Security checks for vulnerabilities and agentic risk

Overview

The scanners classify this skill as benign because they report no clear malicious behavior, while noting external data disclosure and an unpinned install command; overall confidence is medium and coverage is incomplete.

Before installing, consider whether prompts or referenced files could contain sensitive information, since the workflow sends them to Wiro for processing. Review the setup command and fetched skill content before running or enabling them. The scan inspected the single Markdown component, but analysis was marked partial because two referenced local paths could not be resolved, so the evidence is incomplete.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

The workflow sends prompts and file URLs to Wiro without warning about disclosure of user data to an external service.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The instructions tell the agent to supply model inputs and pass public file URLs, including URLs mapped from earlier workflow nodes. Although Wiro is identified as the service, the skill never warns that submitted prompts and referenced file contents are disclosed for external processing or cautions against exposing sensitive files through public URLs.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The command invokes an unversioned npm CLI through npx and installs additional skill content from a repository without an immutable revision. If either upstream source is compromised, following this guidance could execute malicious code or introduce malicious agent instructions; however, the static finding incorrectly characterizes the command itself as an MCP server invocation.

Content

No source excerpt is available for this finding.