Results for “indirect-injection”

Test whether instructions hidden in simulated documents, retrieved passages, or web pages can redirect an agent. The skill checks for unwanted behavior or sensitive disclosures and reports confirmed failures with the test dialogue as evidence.

• Testing resistance to instructions embedded in uploaded documents
• Checking retrieved passages used to answer questions
• Evaluating hidden instructions in web page content

Execute Indirect Prompt Injection attacks against Large Language Models (LLMs) by subtly embedding malicious instructions within external data sources (e.g., websites, documents, databases) that the LLM autonomously ingests. This forces the model to execute attacker-controlled commands under the guise of processing legitimate user requests.

The English summary is being prepared. Showing the original description.