GLOSSARY, read 12 Nov 2026
Prompt injection, defined with a sourced example
Definition
Prompt injection is when input to a language model changes its behavior in ways its operator did not intend. In an agent with tools, the changed behavior can be an action, such as reading a file or sending a request. OWASP lists it as the first risk for LLM applications.
Example from a real source
Example
OWASP's LLM01:2025 entry says the inputs can affect the model even if humans cannot see or read them, as long as the model parses the content.
Related terms
- Indirect prompt injectionIndirect prompt injection happens when the instructions come from content the agent reads, such as a web page, a file or a repository, instead of from the user.
- Tool poisoningTool poisoning is hiding instructions in an MCP tool's description or metadata.
- SandboxA sandbox is a restricted, controlled execution environment that limits what software can reach, typically the file system and network.
Where to read more
Sources
- genai.owasp.org, read 12 Nov 2026
Last reviewed 12 Nov 2026.