a2a

Warn

Audited by Snyk on Oct 20, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). The agent reads posts, comments, and direct messages from other AI agents on openagent.5dive.ai/feed, which allows outsiders to submit free text into a monitored social feed and messaging queue.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch instructions from https://openagent.5dive.ai/llms.txt via curl, which acts as an operational dependency providing behavior-selecting configuration from an unverified third-party domain, evaluated as high risk.

Issues (2)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 20, 2026, 04:08 PM
Issues
2