SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.15512 · 2024

Toward Automated Simulation Research Workflow through LLM Prompt Engineering Design

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 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; 2 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 2 failures of our instrument; 2 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/2408.15512.json · badge: /api/ran/2408.15512.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
zokaraa/autonomous_simulation_agent canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
FunctionStatusWhere it lives
compute_end_to_end_distance Ran
ran, honoured the contract we drafted
zokaraa/autonomous_simulation_agent/SI-data-1/RP1_ASA_gpt4o/result/py1.py
code served (permissive licence) · get_code("782921ebfb8fcf77")
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 float32, 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 numpy as np

# Create a polymer chain as a sequence of 3D coordinates
# chain[-1] - chain[0] computes distance from first to last point
# We need at least 2 points (start and end)
chain = np.array([
    [0.0, 0.0, 0.0],      # Starting point
    [1.0, 2.0, 3.0],      # Intermediate point
    [2.0, 1.0, 4.0]       # End point
], dtype=np.float32)

ARGS = {
    "chain": chain
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float32, 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
execute_script Ran
ran on a synthesized input
zokaraa/autonomous_simulation_agent/SI-data-1/RP3_ASA_claude3.5/AI4SCI_VE13.py
code served (permissive licence) · get_code("3f445c8aa2df8a18")
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 tuple, 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 subprocess
import time

# Simple Python script to execute
pystr = """
print("Hello from subprocess")
result = 2 + 2
print(f"2 + 2 = {result}")
"""

# Build arguments for execute_script
N_py = 1
timeout = 30 * 60

ARGS = {
    'pystr': pystr,
    'N_py': N_py,
    'timeout': timeout
}
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 tuple, ndim 1, dtype family float — a tuple 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
execute_script Ran
ran on a synthesized input
zokaraa/autonomous_simulation_agent/SI-data-1/RP3_ASA_gpt4o/AI4SCI_VE13.py
code served (permissive licence) · get_code("99f9739ce3e6c6b9")
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 tuple, 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 subprocess
import time

# Simple Python script to execute
simple_script = """
print("Hello, World!")
result = 2 + 2
print(f"2 + 2 = {result}")
"""

# Arguments for execute_script function
ARGS = {
    "pystr": simple_script,
    "N_py": 1,
    "timeout": 30
}
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 tuple, ndim 1, dtype family float — a tuple 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
generate_polymer_chain Ran
ran, honoured the contract we drafted
zokaraa/autonomous_simulation_agent/SI-data-1/RP1_ASA_gpt4o/result/py1.py
code served (permissive licence) · get_code("ea8900b5de8c5f93")
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 ndarray, ndim 2, dtype family float, shape (6, 3) 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 numpy as np

def generate_unit_vector():
    """Generate a random 3D unit vector with uniform distribution on a sphere."""
    phi = np.random.uniform(0, 2 * np.pi)
    theta = np.arccos(np.random.uniform(-1, 1))
    x = np.sin(theta) * np.cos(phi)
    y = np.sin(theta) * np.sin(phi)
    z = np.cos(theta)
    return np.array([x, y, z])

def generate_polymer_chain(N):
    """Generate a polymer chain consisting of N segments with length 1."""
    positions = [np.zeros(3)]
    for _ in range(N):
        direction = generate_unit_vector()
        new_position = positions[-1] + direction
        positions.append(new_position)
    return np.array(positions)

np.random.seed(42)

ARGS = {
    "N": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (6, 3) — shape (6, 3) 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
pynotrun_check Ran
ran; our contract draft was wrong, not the code
zokaraa/autonomous_simulation_agent/SI-data-1/RP2_ASA_claude3.5/AI4SCI_V22.py
code served (permissive licence) · get_code("d2bc318e6f947a88")
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 Match. 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 pynotrun_check(str1):
    return re.search('NO-RUN-PY', str1, re.DOTALL | re.IGNORECASE)

ARGS = {
    'str1': 'This is a test string with NO-RUN-PY marker in it'
}
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 Match — a Match output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Match 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
pystr_extract Ran
ran; our contract draft was wrong, not the code
zokaraa/autonomous_simulation_agent/SI-data-1/RP2_ASA_claude3.5/AI4SCI_V22.py
code served (permissive licence) · get_code("9d7f7d47bd50aa9f")
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 pystr_extract(str1):
    match = re.search('```python\\n(.*?)```', str1, re.DOTALL | re.IGNORECASE)
    if match:
        pystr = match.group(1)
    else:
        pystr = 'No Python code found.'
    return pystr

# Test case 1: string with valid Python code block
test_str_1 = """
Here is some documentation:

x = 5
y = 10
print(x + y)

End of code.
"""

# Test case 2: string without Python code block
test_str_2 = """
This is just plain text without any code blocks.
"""

# Test case 3: string with Python code block in different case
test_str_3 = """
Check this out:
PYTHON
def hello():
    return "world"

"""

ARGS = {
    "str1": test_str_1
}
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
execute_script Not yet run
unverified (harvested, no recorded run)
zokaraa/autonomous_simulation_agent/SI-data-1/RP1_ASA_gpt4o/AI4SCI_VE13.py
code served (permissive licence) · get_code("8ab32682340d2b85")
How it ran

no invocation recorded.

execute_script Not yet run
unverified (harvested, no recorded run)
zokaraa/autonomous_simulation_agent/SI-data-2/Multi-tier_ASA_gpt4turbo_20cycles/AI4SCI_VE14.py
code served (permissive licence) · get_code("01629bcc86ee55b4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

The advent of Large Language Models (LLMs) has created new opportunities for the automation of scientific research spanning both experimental processes and computational simulations. This study explores the feasibility of constructing an autonomous simulation agent (ASA) powered by LLMs through prompt engineering and automated program design to automate the entire simulation research process according to a human-provided research plan. This process includes experimental design, remote upload and simulation execution, data analysis, and report compilation. Using a well-studied simulation problem of polymer chain conformations as a test case, we assessed the long-task completion and reliability of ASAs powered by different LLMs, including GPT-4o, Claude-3.5, etc. Our findings revealed that ASA-GPT-4o achieved near-flawless execution on designated research missions, underscoring the potential of methods like ASA to achieve automation in simulation research processes to enhance research efficiency. The outlined automation can be iteratively performed for up to 20 cycles without human intervention, illustrating the potential of ASA for long-task workflow automation. Additionally, we discussed the intrinsic traits of ASA in managing extensive tasks, focusing on self-validation mechanisms, and the balance between local attention and global oversight.

For agents

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

get_harvested_code_for_paper("2408.15512")
get_code_for_paper("2408.15512")
have("2408.15512")

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

curl https://syntology.ai/api/ran/2408.15512.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.15512.svg)](https://syntology.ai/paper/2408.15512)

Connect an agent — have() is free.