SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.08804 · 2024

FinRobot: AI Agent for Equity Research and Valuation with Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 8 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2411.08804.json · badge: /api/ran/2411.08804.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ai4finance-foundation/finrobot canonical (classifier label)
classifier basis: census
10 ran · 1 no recorded run
FunctionStatusWhere it lives
clean_speakers Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/data_source/earnings_calls_src/main_earningsData.py
code served (permissive licence) · get_code("f8b1589c59875d69")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def clean_speakers(speaker):
    speaker = re.sub('\n', '', speaker)
    speaker = re.sub(':', '', speaker)
    return speaker

ARGS = {
    'speaker': 'Speaker Name\nWith:\nColons'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
correct_date Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/data_source/earnings_calls_src/earningsData.py
code served (permissive licence) · get_code("5ff47cb0f312d6ba")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

from datetime import datetime

ARGS = {
    'yr': 2023,
    'dt': '2022-05-15 14:30:45'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
decorate_all_methods Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/utils.py
code served (permissive licence) · get_code("8e1541083f520990")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import functools

def simple_decorator(func):
    """A simple decorator that wraps a function."""
    @functools.wraps(func)
    def wrapper(*args, **kwargs):
        return func(*args, **kwargs)
    return wrapper

class SampleClass:
    """A sample class with various callable and non-callable attributes."""
    def method_one(self):
        return "method_one"
    
    def method_two(self, x):
        return x * 2
    
    @staticmethod
    def static_method():
        return "static"
    
    @classmethod
    def class_method(cls):
        return "class_method"
    
    non_callable_attr = 42

ARGS = {
    "decorator": simple_decorator
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_speakers Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/data_source/earnings_calls_src/earningsData.py
code served (permissive licence) · get_code("4e21ffca58abf22f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re
from typing import List

def extract_speakers(cont: str) -> List[str]:
    """Extract the list of speakers

    Args:
        cont (str): transcript content

    Returns:
        List[str]: list of speakers
    """
    pattern = re.compile('\\n(.*?):')
    matches = pattern.findall(cont)
    return list(set(matches))


# Setup code to construct valid arguments
transcript_content = """
Alice: Hello, how are you today?
Bob: I'm doing great, thanks for asking!
Alice: That's wonderful to hear.
Charlie: Mind if I join the conversation?
Bob: Of course, the more the merrier!
"""

ARGS = {
    "cont": transcript_content
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_next_weekday Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/utils.py
code served (permissive licence) · get_code("f99e5a772c6ec719")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned datetime. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

from datetime import date, timedelta, datetime

# The function expects a 'date' parameter which can be either:
# 1. A datetime object
# 2. A string in format '%Y-%m-%d'

# Test case 1: Pass a datetime object that is a weekend (Saturday)
test_date = datetime(2024, 1, 6)  # This is a Saturday

ARGS = {
    "date": test_date
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type datetime — a datetime output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a datetime output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_regex_enum Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/data_source/filings_src/sec_filings.py
code served (permissive licence) · get_code("2f2dac555c9b83c3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned CustomSECSection. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re
from enum import Enum

def get_regex_enum(section_regex):
    """Get sections using regular expression

    Args:
        section_regex (str): regular expression for the section name

    Returns:
        CustomSECSection.CUSTOM: Custom regex section name
    """

    class CustomSECSection(Enum):
        CUSTOM = re.compile(section_regex)

        @property
        def pattern(self):
            return self.value
    return CustomSECSection.CUSTOM


ARGS = {
    "section_regex": r"^SEC_\w+$"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CustomSECSection — a CustomSECSection output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CustomSECSection output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
instruction_message Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/agents/utils.py
code served (permissive licence) · get_code("ca8ad47a235cd956")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

class MockRecipient:
    def __init__(self, instruction_content):
        self.instruction_content = instruction_content
    
    def chat_messages_for_summary(self, sender):
        return [
            {'content': f'instruction & resources saved to {self.instruction_content}'}
        ]

class MockSender:
    pass

class MockConfig:
    pass

# Create a temporary file with instruction content
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
instruction_text = "This is the instruction content.\nPlease follow these steps."
temp_file.write(instruction_text)
temp_file.close()
temp_file_path = temp_file.name

# Construct arguments
recipient = MockRecipient(temp_file_path)
messages = []
sender = MockSender()
config = MockConfig()

ARGS = {
    'recipient': recipient,
    'messages': messages,
    'sender': sender,
    'config': config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
instruction_trigger Ran
ran, violated the contract we drafted
ai4finance-foundation/finrobot/finrobot/agents/utils.py
code served (permissive licence) · get_code("f578758cc180c855")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

class MockSender:
    def __init__(self, message_content):
        self._message_content = message_content
    
    def last_message(self):
        return {'content': self._message_content}

# Test case 1: message contains the trigger phrase
sender_with_trigger = MockSender('instruction & resources saved to /path/to/file')

ARGS = {
    'sender': sender_with_trigger
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
order_trigger Ran
ran, violated the contract we drafted
ai4finance-foundation/finrobot/finrobot/agents/utils.py
code served (permissive licence) · get_code("37f24fa58d0b72be")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

class MockSender:
    def __init__(self, name, last_message_content):
        self.name = name
        self._last_message_content = last_message_content
    
    def last_message(self):
        return {'content': self._last_message_content}

sender = MockSender(name="OrderBot", last_message_content="Please order pizza")
name = "OrderBot"
pattern = "order"

ARGS = {
    "sender": sender,
    "name": name,
    "pattern": pattern
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
stringify_output Ran
ran; our contract draft was wrong, not the code
ai4finance-foundation/finrobot/finrobot/toolkits.py
code served (permissive licence) · get_code("6bc3e33e9fc97e10")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from pandas import DataFrame
from functools import wraps

def stringify_output(func):
    @wraps(func)
    def wrapper(*args, **kwargs):
        result = func(*args, **kwargs)
        if isinstance(result, DataFrame):
            return result.to_string()
        else:
            return str(result)
    return wrapper

# Create a sample function to decorate
def sample_function_returning_dataframe():
    return DataFrame({'a': [1, 2], 'b': [3, 4]})

def sample_function_returning_string():
    return "hello"

def sample_function_returning_number():
    return 42

# Test the decorator with different return types
decorated_df_func = stringify_output(sample_function_returning_dataframe)
decorated_str_func = stringify_output(sample_function_returning_string)
decorated_num_func = stringify_output(sample_function_returning_number)

# Build ARGS - the stringify_output function takes 'func' as parameter
ARGS = {
    'func': sample_function_returning_dataframe
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_earnings_transcript Not yet run
unverified (harvested, no recorded run)
ai4finance-foundation/finrobot/finrobot/data_source/earnings_calls_src/earningsData.py
code served (permissive licence) · get_code("adb80ec201ab052c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 11 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

As financial markets grow increasingly complex, there is a rising need for automated tools that can effectively assist human analysts in equity research, particularly within sell-side research. While Generative AI (GenAI) has attracted significant attention in this field, existing AI solutions often fall short due to their narrow focus on technical factors and limited capacity for discretionary judgment. These limitations hinder their ability to adapt to new data in real-time and accurately assess risks, which diminishes their practical value for investors. This paper presents FinRobot, the first AI agent framework specifically designed for equity research. FinRobot employs a multi-agent Chain of Thought (CoT) system, integrating both quantitative and qualitative analyses to emulate the comprehensive reasoning of a human analyst. The system is structured around three specialized agents: the Data-CoT Agent, which aggregates diverse data sources for robust financial integration; the Concept-CoT Agent, which mimics an analysts reasoning to generate actionable insights; and the Thesis-CoT Agent, which synthesizes these insights into a coherent investment thesis and report. FinRobot provides thorough company analysis supported by precise numerical data, industry-appropriate valuation metrics, and realistic risk assessments. Its dynamically updatable data pipeline ensures that research remains timely and relevant, adapting seamlessly to new financial information. Unlike existing automated research tools, such as CapitalCube and Wright Reports, FinRobot delivers insights comparable to those produced by major brokerage firms and fundamental research vendors. We open-source FinRobot at \url{https://github. com/AI4Finance-Foundation/FinRobot}.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2411.08804")
get_code_for_paper("2411.08804")
have("2411.08804")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2411.08804.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2411.08804.svg)](https://syntology.ai/paper/2411.08804)

Connect an agent — have() is free.