TinyAgentsTemplatesInvestment Analyst
Investment Analyst · agent template

Every investment question means an afternoon of hunting across a database, a stack of PDFs, and a spreadsheet nobody updated.

Your whole research desk, working as one agent team

A manager agent routes each research question to the right specialist: one for fund numbers, one for filings, one for the quant work. Every answer comes back written, compared, and cited so you can act on it.

A grounded report in minutes5 agentsNo-code fork
investment-research.rundone
Research Leadplan

Splits the question: returns, filings, risk

Numbers Analysttool

Fund A 3yr 8.2%, Fund B 6.1%

Document Readertool

Flagged liquidity note in latest filing

Cited comparisonanswer

Return gap, risk flag, source per claim

one run · manager plans, specialists report, you get one answer

The old way

The job, before you had a team for it.

This is the work Investment Analyst quietly takes off your plate.

Answers live in a database, a PDF, and a spreadsheet at once

Senior analysts burn hours on junior gathering work

Copy-paste between tools invites errors into client-facing work

A plain chatbot sounds confident but cannot show its sources

The team · 5 agents

A team of agents, not one prompt.

A manager agent reads the goal and hands each part to a specialist built for it, then reconciles their work into one result.

Manager agent
Research Lead

Reads your question, routes each part to the right specialist, and merges the findings into one cited answer

01 · specialist
Numbers Analyst

Queries your fund database for NAV, AUM, and returns, then hands back clean figures with a short summary.

02 · specialist
Document Reader

Scans filings, meeting notes, and fund documents to pull the relevant passage and cite exactly where it came from.

03 · specialist
Metrics Analyst

Runs the quant work: performance calculations, risk metrics, and attribution, and flags anything a PM should see.

04 · specialist
Fact Checker

Checks every number and quote against its source before the answer leaves, so nothing ungrounded slips through.

The run

How it runs.

Fork it, point it at your inputs, and let the team work.

STEP 01
Connect your data

Link your fund database and upload your filings and notes.

STEP 02
Ask in plain English

Type the question the way you would ask a colleague.

STEP 03
Agents divide the work

The manager routes each part to the numbers, document, and metrics specialists.

STEP 04
Read the grounded answer

Get one written response with a source attached to every claim.

See a real run

Compare two credit funds and flag filing risk.

You ask

"Compare our two credit funds on trailing 3-year returns and flag anything in the latest filings that worries you."

the team gets to work
You get back
  • 3-year returns and AUM pulled for both funds
  • A calculated return gap and basic risk comparison
  • Quoted risk language from each fund's latest filing
  • One written answer with a source cited on every figure

Where teams put it to work.

01
Equity analyst

Compresses a half-day of gathering into a few minutes so more time goes to the actual call.

02
Portfolio manager

Gets a cited fund comparison and risk flags before a committee meeting.

03
Founder or investor

Screens funds and filings without hiring a full research desk.

What you provide
  • Access to your fund database (NAV, AUM, returns)
  • Filings, meeting notes, and fund documents to read
  • A plain-English research question
What you get back
  • A written answer to your question
  • A source cited for every number and quote
  • Calculated metrics like returns, risk, and attribution
  • Flags on anything a portfolio manager should review
The basics

What is an investment research tool?

An investment research tool gathers, analyzes, and summarizes financial data so you decide faster. This one uses a team of AI agents that query numbers, read filings, and run risk math, then merge it into one cited answer.

The old version was a data terminal plus a spreadsheet plus a stack of PDFs. This version does the connecting work for you.

Why use an agent team instead of one AI chatbot?

A single chatbot is mediocre at data, documents, and math all at once, and often cannot cite its sources. A manager-plus-specialists layout gives each job to a focused agent, which is more accurate and fully grounded.

It mirrors how real desks are staffed: a data person, a documents person, and a quant, coordinated by a lead.

New to this? Read our guide on how to build a sales battlecard.

Questions, answered.

Can I trust the numbers it gives me?

Yes, when you use the citations. Every figure and quote links back to its source, and a fact-checking step verifies claims before the answer leaves. Treat the output as a strong first draft and review it, as regulators expect a human in the loop.

Do I need a data science team to set it up?

No. You start from the template, connect your database, and upload your documents. There is no model training or engineering project involved.

How is this different from a regular AI chatbot?

A chatbot tries to answer everything alone and often cannot cite its sources. This tool splits the work across specialist agents and grounds every claim in your actual data and documents.

Start here

Fork Investment Analyst and make it yours.

Start from a working team of agents, then swap the tools, tune the instructions, and run it in your own workflows.

Fork this template