Today, we are introducing Specter Agent, the AI agent built for private markets.

Specter Agent helps you find the right opportunities sooner, understand companies and markets faster, and walk into every conversation with the context you need.

Ask it a question the way you would ask a colleague, and Specter Agent will take it from there. It works across Specter's private-market intelligence and the web, completes the underlying research, and returns a sourced answer you can continue working with.

It helps your team make better decisions, faster.

A new way to put Specter's intelligence to work

Ask it to source companies against a thesis, find people emerging from relevant teams, identify investors showing interest, or follow funding and M&A activity. Describe the criteria in plain language, then refine the result as the conversation develops.

Specter Agent can search across its databases, build Saved Searches, compare and prioritize the results, and turn the companies that matter into Lists and Folders. Bring specific entities or existing workspace research into the conversation when you want the Agent to continue from what your team already knows.

Web search adds current external context, while sources keep the result connected to the information behind it.

Find the opportunity and understand why it matters

Specter Agent expands how much of the market your team can investigate. More theses can be tested, more signals can be followed, and more promising leads can receive deeper research.

It can then help explain why a company, person, or signal matters. Ask it to assess an opportunity, compare it with the market, identify the strongest indicators, understand what has changed, and surface the potential upside or risk.

The Agent can combine funding and transaction history, revenue and valuation estimates, team movements, investor interest, competitive context, and current web research. That wider context helps your team move beyond finding information and toward deciding which opportunities deserve action.

Make it work the way your team works

Every team has its own way of sourcing companies, screening opportunities, researching markets, and preparing for decisions.

With Skills, you can teach Specter Agent how your team approaches that work. Document the process once, share it with the workspace, and improve it over time. The whole team can benefit from the same sourcing method, company brief, or market review without rebuilding it each time.

Personalization gives the Agent context about the team's preferred markets, stages, criteria, and output. It can also continue from the companies, people, investors, Lists, Saved Searches, and other research already inside the workspace.

Together, these capabilities allow Specter Agent to work from both the team's process and the knowledge it has already built.

An agent is only as good as the intelligence behind it

A general-purpose model can retrieve and summarize information that is already publicly available. Specter Agent can go further because it works directly with Specter's proprietary records, estimates, and signals.

It is built on more than 8 million companies, 620 million people, 340,000 investors, and one million transactions, together with Talent Signals, Interest Signals, Revenue Signals, company performance estimates, and detailed profiles.

This data layer connects past performance and transactions with present momentum and early indicators of where companies, investors, and markets may be moving.

That allows Specter Agent to help answer more consequential questions:

Which companies are building momentum?
Where is investor attention forming?
Which opportunities look strongest against the team's criteria?

By bringing those layers together, Specter Agent can help find, compare, and prioritize opportunities across an investment workflow. It is not limited to telling you what is already known. It helps build a forward-looking view of what may matter next.

Available today, with more coming next

Today, Specter Agent can search, research, build Saved Searches, create Lists and Folders, work with tagged entities, search the web, follow personalized instructions, and run repeatable Skills.

Next, scheduled tasks will allow it to run recurring work in the background. Memory will help it retain useful context across conversations. Context-aware chat will bring the Agent into company and people profiles and across more of Specter, ready to work from the page and information already in front of you.

We want to make Specter Agent the intelligence layer that works wherever you and your team work.