ai-taking-actions
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The skill explicitly configures and uses web-search and scraper tools (e.g., LangChain DuckDuckGoSearchRun, WikipediaQueryRun, wttr.in in get_weather, and dspy.ColBERTv2 over wiki17_abstracts) so the agent fetches and ingests open/public third‑party (including user‑generated wiki) content and then reads/interprets those results as part of its workflow, enabling indirect prompt injection.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill calls dspy.ColBERTv2(url="http://20.102.90.50:2017/wiki17_abstracts") at runtime to fetch text that is injected into tool outputs and the model context (e.g., search functions), so remote content from that URL can directly influence/drive agent prompts and behavior.
Audit Metadata