Introducing the Chat Driven Symbolic Agent

We are excited this week to announce the first step of our plan for a chat driven universal Symbolic agent that can drive all aspects of our platform. The rollout begins in Symbolic Research, where users can now request nearly any work process simply by asking for it. The agent will walk the user through its steps to execute the task, and ask for any guidance it needs along the way.

We believe that the combination of Symbolics' smart workspace with a chat driven agent is unique and powerful. Teams can collaborate and organize information in the workspace while using the agent as a virtual member of the project team. This sets up both new and experienced users to deliver consistently top tier professional quality content with high productivity.

Here's what's new in Symbolic:

Agentic Chat In Research

The Symbolic agent can now be called through the "Ask Symbolic" button at the top of the Research space. Using conversational chat, the user can request any function by simply asking for it (see below for examples). Chat history and its context is fully preserved within any project. If a user requires a new chat session, or their entire history of chat across all projects, they can access that through the icons in the top menu.

Find Reference Material Anywhere

Request any research materials through chat. Examples of this are "find me documents in my Library related to the topic", "find me documents on the web that explain...", "find me articles from anywhere that contradict my thesis". Results will be presented in the chat window, and can then be added to Research space.

Tag and Organize Research

Ask the agent to automatically tag articles in the research space according to any criteria (topic, length, relevance to the project). Those tags can then be used to organize the research and target it to specific sections in drafts such as newsletters.

Create Custom Research Requests

Any custom research question or action can be executed by the Symbolic agent. "Summarize the first two articles, and pull quotes from the third"; "answer these 3 questions about the research..."; "extract the data about IPO performance but exclude all other data". Custom research requests can relate to both text and data based files.

Execute Complex Multi-Part Jobs

The agent can execute complex multi-part workflows. As an example, you can ask Symbolic to find documents related to xyz, summarize them and then tag each one according to its specific sub topic.

That's it from us this week. We are always working to improve Symbolic and make your experience more efficient and enjoyable. We would love to hear your feedback and any suggestions you may have for future updates. Don't hesitate to reach out to team@symbolic.ai.