[onechat] Implement agent modes#224801
Conversation
|
/ci |
|
/ci |
|
/ci |
|
/ci |
|
/ci |
|
|
||
| const setResearchGoalTool = () => { | ||
| return new DynamicStructuredTool({ | ||
| name: setResearchGoalToolName, |
There was a problem hiding this comment.
very nitty but was this suppose to be researchGoalToolName instead?
There was a problem hiding this comment.
Tool's name is set_research_goal, it's the "internal" tool exposed to the LLM in the first step to define the research goal based on the conversation. It's exposed as a tool because the LLM can also decide to answer with aplain text response if it thinks it need more info before starting the deep-research process
💛 Build succeeded, but was flaky
Failed CI StepsTest Failures
Metrics [docs]Public APIs missing comments
Public APIs missing exports
Page load bundle
Unknown metric groupsAPI count
History
|
researcher agent mode
Summary
Follow-up of #223367
Fix https://github.com/elastic/search-team/issues/10259
This PR introduce the concept of agent mode, expose the
researchagent as a mode instead of a tool, and add thereasonandplanmodes.Examples
Calling the Q/A (default) mode
Calling the researcher mode