v1.12.0
What's Changed
- Augment prompt injection protection by @stephenyeargin
Added a sanitization layer in src/utils/ollama-utils.js to detect and neutralize common prompt injection patterns before they reach the model. The protection is applied inside src/hubot-ollama.js at the point where user input is assembled into the message chain. New test coverage added in test/ollama-utils.test.js.
Full Changelog: v1.11.2...v1.12.0