SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2304.03442 · 2023

Generative Agents: Interactive Simulacra of Human Behavior

Percy Liang, Michael Bernstein, Meredith Morris, Joon Park, Carrie Cai, Joseph O'brien

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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; 1 honoured the contract we drafted, 1 violated it and 5 have no contract verdict; 3 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/2304.03442.json · badge: /api/ran/2304.03442.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
joonspk-research/generative_agents canonical (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
renmengjie7/aisimutoolkit community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
litanlitudan/skyagi community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
jina-ai/thinkgpt community (classifier label)
classifier basis not recorded on the link
1 ran · 2 no recorded run
FunctionStatusWhere it lives
create_folder_if_not_there Ran
ran, violated the contract we drafted
joonspk-research/generative_agents/environment/frontend_server/global_methods.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from sxswz213/crsec, is pointer only (licence: NONE) · get_code("04d3df25d420010e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from sxswz213/crsec, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

format_options Ran
ran on a synthesized input
jina-ai/thinkgpt/thinkgpt/gpt_select.py
code served (permissive licence) · get_code("d52b33b77d35e4ed")
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 torch

torch.manual_seed(0)

ARGS = {
    "options": ["option1", "option2", "option3"]
}
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 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
get_file_stream Ran
ran on a synthesized input
renmengjie7/aisimutoolkit/AISimuToolKit/utils/utils.py
code served (permissive licence) · get_code("2793537d25115ebc")
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 BufferedReader. 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

# Create a temporary file for testing
temp_file = tempfile.NamedTemporaryFile(mode='wb', delete=False, suffix='.bin')
temp_file.write(b'test data for finetune')
temp_file.close()
temp_file_path = temp_file.name

ARGS = {
    'file': temp_file_path
}
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 BufferedReader — a BufferedReader output has no array shape of its own; none is 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_fromat_len Ran
ran on a synthesized input
renmengjie7/aisimutoolkit/AISimuToolKit/utils/utils.py
code served (permissive licence) · get_code("56a62d76a2e1fb94")
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 int, ndim 0, dtype family int, 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

torch.manual_seed(0)

# The function get_fromat_len takes a single integer parameter 'num'
# and returns the maximum of 2 or (num // 10 + 1)
# This is used for naming specifications to ensure uniform formatting

ARGS = {
    "num": 42
}
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 int, ndim 0, dtype family int, 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
parse_yaml_config Ran
ran on a synthesized input
renmengjie7/aisimutoolkit/AISimuToolKit/utils/utils.py
code served (permissive licence) · get_code("ff183b964718e88e")
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 dict. 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
import yaml

# Create a temporary YAML configuration file
temp_dir = tempfile.mkdtemp()
config_path = os.path.join(temp_dir, "agent_config.yaml")

# Create a valid YAML configuration
config_data = {
    "agent": "dqn",
    "learning_rate": 0.001,
    "gamma": 0.99,
    "epsilon": 0.1,
    "hidden_size": 128,
    "layers": [64, 128, 256],
    "activation": "relu"
}

with open(config_path, 'w', encoding='utf-8') as f:
    yaml.dump(config_data, f)

ARGS = {
    "path": config_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
read_file_to_list Ran
ran; our contract draft was wrong, not the code
joonspk-research/generative_agents/environment/frontend_server/global_methods.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from sxswz213/crsec, is pointer only (licence: NONE) · get_code("48476b17bdfd0043")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from sxswz213/crsec, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

read_file_to_set Ran
ran; our contract draft was wrong, not the code
joonspk-research/generative_agents/environment/frontend_server/global_methods.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from sxswz213/crsec, is pointer only (licence: NONE) · get_code("219bfa7864169c24")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from sxswz213/crsec, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

relevance_score_fn Ran
ran, honoured the contract we drafted
litanlitudan/skyagi/skyagi/src/skyagi/simulation/simulation.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from cr-gjx/suspicion-agent, is pointer only (licence: NONE) · get_code("545ba72b86f890f5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from cr-gjx/suspicion-agent, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

verify_openai_ Ran
ran on a synthesized input
litanlitudan/skyagi/skyagi/src/skyagi/util.py
code served (permissive licence) · get_code("f1c5580780f0e534")
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 os

# The function requires a valid OpenAI API token string
# Since we're in a sandbox without actual network access,
# we'll provide a properly formatted but fake token
# The function will attempt to use it but fail at the API call level
# rather than at the argument construction level

ARGS = {
    'token': 'sk-' + 'a' * 48  # Fake but properly formatted OpenAI API key
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
openai — 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
verify_pinecone_token Ran
ran; our contract draft was wrong, not the code
litanlitudan/skyagi/skyagi/src/skyagi/util.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from cr-gjx/suspicion-agent, is pointer only (licence: NONE) · get_code("2d779390d2d954c2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from cr-gjx/suspicion-agent, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

fit_context Not yet run
unverified (harvested, no recorded run)
jina-ai/thinkgpt/thinkgpt/helper.py
code served (permissive licence) · get_code("ea817e6b698cf08e")
How it ran

no invocation recorded.

get_n_tokens Not yet run
unverified (harvested, no recorded run)
jina-ai/thinkgpt/thinkgpt/helper.py
code served (permissive licence) · get_code("4c2c52a526fb9a29")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 5 later) and 7 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.

FunctionAlso present inEarliest-datedOthers named
create_folder_if_not_there 1 other paper
1 later
2403.08251 (2024-03)
Emergence of Social Norms in Generative Agent Societies: Principles and Architecture
read_file_to_list 1 other paper
1 later
2403.08251 (2024-03)
Emergence of Social Norms in Generative Agent Societies: Principles and Architecture
read_file_to_set 1 other paper
1 later
2403.08251 (2024-03)
Emergence of Social Norms in Generative Agent Societies: Principles and Architecture
relevance_score_fn 1 other paper
1 later
2309.17277 (2023-09)
Suspicion-Agent: Playing Imperfect Information Games with Theory of Mind Aware GPT-4
verify_pinecone_token 1 other paper
1 later
2309.17277 (2023-09)
Suspicion-Agent: Playing Imperfect Information Games with Theory of Mind Aware GPT-4

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

Figure 1: Generative agents are believable simulacra of human behavior for interactive applications. In this work, we demonstrate generative agents by populating a sandbox environment, reminiscent of The Sims, with twenty-five agents. Users can observe and intervene as agents plan their days, share news, form relationships, and coordinate group activities.

For agents

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

get_harvested_code_for_paper("2304.03442")
get_code_for_paper("2304.03442")
have("2304.03442")

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

curl https://syntology.ai/api/ran/2304.03442.json

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

[![Syntology run record](https://syntology.ai/api/ran/2304.03442.svg)](https://syntology.ai/paper/2304.03442)

Connect an agent — have() is free.