SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.07002 · 2024

CursorCore: Assist Programming through Aligning Anything

arXiv · PDF · Open in the Atlas

Code that ran

13 of 13 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; 4 honoured the contract we drafted, 3 violated it and 0 have no contract verdict; 6 failures of our instrument; 0 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/2410.07002.json · badge: /api/ran/2410.07002.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
TechxGenus/CursorCore canonical (classifier label)
classifier basis: census
13 ran · 0 no recorded run
FunctionStatusWhere it lives
blockwise_if_continuous_modify Ran
ran, violated the contract we drafted
TechxGenus/CursorCore/src/utils.py
code served (permissive licence) · get_code("21cc834ebb3bd7ef")
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 difflib
import Levenshtein

# Simple test cases for the blockwise_if_continuous_modify function
code1 = """def hello():
    print("hello")
    return 1"""

code2 = """def hello():
    print("hello world")
    return 1"""

code3 = """def hello():
    print("hello world")
    return 2"""

ARGS = {
    "code1": code1,
    "code2": code2,
    "code3": code3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
Levenshtein — 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
generate_diff Ran
ran; our contract draft was wrong, not the code
TechxGenus/CursorCore/src/utils.py
code served (permissive licence) · get_code("21907cbb784ea33c")
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 difflib

# Construct setup code for generate_diff function
code1 = """def hello():
    print("Hello")
    return 42"""

code2 = """def hello():
    print("Hello, World!")
    return 43"""

fromfile = "original.py"
tofile = "modified.py"
n = 3

ARGS = {
    'code1': code1,
    'code2': code2,
    'fromfile': fromfile,
    'tofile': tofile,
    'n': n
}
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
get_changes Ran
ran, honoured the contract we drafted
TechxGenus/CursorCore/src/post_collection.py
code served (permissive licence) · get_code("e43ddd2a28d90d8e")
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 2, dtype family int; 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 Levenshtein

# Test cases for get_changes function
s1_simple = "kitten"
s2_simple = "sitting"

s1_identical = "hello"
s2_identical = "hello"

s1_empty = ""
s2_empty = ""

s1_insertion = "abc"
s2_insertion = "aXbc"

s1_deletion = "abcd"
s2_deletion = "acd"

s1_substitution = "cat"
s2_substitution = "bat"

# ARGS dict with all parameters for get_changes
ARGS = {
    "s1": "kitten",
    "s2": "sitting"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
Levenshtein — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family int — 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
if_continuous_modify Ran
ran, violated the contract we drafted
TechxGenus/CursorCore/src/utils.py
code served (permissive licence) · get_code("57b8a8ace06f9a19")
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 Levenshtein

# The function takes three string parameters representing code
# It uses Levenshtein.distance to compute edit distances between pairs
# No models, tensors, or complex objects needed - just strings

ARGS = {
    "code1": "hello world",
    "code2": "hello wor1d",
    "code3": "hello wor1d extra"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
Levenshtein — 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
line_cost Ran
ran, honoured the contract we drafted
TechxGenus/CursorCore/eval/search_and_replace.py
code served (permissive licence) · get_code("916956e5856475da")
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 float, 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

ARGS = {
    'line': 'def hello_world():'
}
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 float, 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
match_without_whitespace Ran
ran, violated the contract we drafted
TechxGenus/CursorCore/eval/search_and_replace.py
code served (permissive licence) · get_code("980c046abbc74a62")
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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "str1": "  hello world  ",
    "str2": "hello world"
}
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
postprocess_output_lc Ran
ran; our contract draft was wrong, not the code
TechxGenus/CursorCore/eval/utils.py
code served (permissive licence) · get_code("b5c86200852e285c")
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.

# First, define the NEXT_START and NEXT_END markers that the function expects
NEXT_START = "NEXT_START"
NEXT_END = "NEXT_END"

# Create sample arguments for the postprocess_output_lc function
current_code = """def hello():
    print("world")
    x = 1
    y = 2
    return x + y"""

# Create an output string with the expected format containing code modifications
output = f"""{NEXT_START}
1,2

    print("modified")
    x = 10

{NEXT_END}"""

ARGS = {
    "current": current_code,
    "output": output,
}
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
postprocess_output_sr Ran
ran; our contract draft was wrong, not the code
TechxGenus/CursorCore/eval/utils.py
code served (permissive licence) · get_code("66a98652392c56f7")
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

# Define the sentinel markers and helper function that the code uses
NEXT_START = "<<<NEXT_START>>>"
NEXT_END = "<<<NEXT_END>>>"
SEARCH_AND_REPLACE = "<<<SEARCH_AND_REPLACE>>>"

class MatchResult:
    """Stand-in class for the result of find_best_match"""
    def __init__(self, start, end):
        self.start = start
        self.end = end

def find_best_match(before, current):
    """Stand-in implementation of find_best_match"""
    # Simple implementation: find the before string in current and return match indices
    lines = current.split('\n')
    start_idx = 0
    end_idx = len(lines)
    
    # Try to find the before pattern in the current text
    before_lines = before.split('\n')
    for i in range(len(lines) - len(before_lines) + 1):
        if lines[i:i+len(before_lines)] == before_lines:
            start_idx = i
            end_idx = i + len(before_lines)
            break
    
    return MatchResult(start=start_idx, end=end_idx)

# Construct valid arguments for postprocess_output_sr
current_str = """def hello():
    print("world")
    return 42"""

output_str = f"""{NEXT_START}

def hello():
    print("world")
    return 42
{SEARCH_AND_REPLACE}
def hello():
    print("goodbye")
    return 43

{NEXT_END}"""

ARGS = {
    "current": current_str,
    "output": output_str
}
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
postprocess_output_wf Ran
ran; our contract draft was wrong, not the code
TechxGenus/CursorCore/eval/utils.py
code served (permissive licence) · get_code("28bc4376d8a33de9")
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

# Define the markers used in the function
NEXT_START = "[WORKFLOW_START]"
NEXT_END = "[WORKFLOW_END]"

# Create a sample output string that contains the expected markers and pattern
current_str = "fallback_value"

output_str = """Some prefix text
[WORKFLOW_START]
Here is some content

print("hello")

[WORKFLOW_END]
Some suffix text"""

ARGS = {
    "current": current_str,
    "output": output_str
}
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
ratio_str Ran
ran; our contract draft was wrong, not the code
TechxGenus/CursorCore/eval/extract_results.py
code served (permissive licence) · get_code("2f5254e21188f550")
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)

# ratio_str takes two numeric arguments: numerator and denominator
# The function performs arithmetic operations on them (division, comparison to 0)
# and formats them as strings

# Test case 1: normal case with non-zero denominator
numerator_1 = 3
denominator_1 = 4

# Test case 2: zero denominator case
numerator_2 = 5
denominator_2 = 0

# We'll use the first test case for ARGS
ARGS = {
    "numerator": 3,
    "denominator": 4
}
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
sample_from_boundaries Ran
ran, honoured the contract we drafted
TechxGenus/CursorCore/src/post_collection.py
code served (permissive licence) · get_code("f8e59191046eada4")
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 int; 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 random

random.seed(0)

ARGS = {
    "left": 5,
    "right": 20,
    "max_length": 50,
    "noisy_length": 3
}
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 int — 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
sample_from_ranges Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TechxGenus/CursorCore/src/post_collection.py
code served (permissive licence) · get_code("a6ac45fd74c40523")
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 random

def sample_from_ranges(ranges):
    """
    Selects a random number from a set of ranges.

    Args:
        ranges (list of tuple): A list of tuples where each tuple contains two integers (start, end) 
                                representing the inclusive range from start to end.

    Returns:
        int: A randomly selected number from the combined ranges.

    Raises:
        ValueError: If the ranges list is empty or if no numbers can be generated from the given ranges.
    """
    numbers = set()
    for start, end in ranges:
        numbers.update(range(start, end + 1))
    return random.choice(list(numbers))


ARGS = {
    "ranges": [(1, 5), (10, 15), (20, 22)]
}
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 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
score_line Ran
ran, honoured the contract we drafted
TechxGenus/CursorCore/eval/search_and_replace.py
code served (permissive licence) · get_code("9a30f2d2fabf7319")
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 float, 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.

from functools import lru_cache
from rapidfuzz import fuzz

@lru_cache()
def score_line(str1: str, str2: str) -> float:
    if str1 == str2:
        return 100
    if str1.lstrip() == str2.lstrip():
        whitespace_ratio = abs(len(str1) - len(str2)) / (len(str1) + len(str2))
        score = 90 - whitespace_ratio * 10
        return max(score, 0)
    if str1.strip() == str2.strip():
        whitespace_ratio = abs(len(str1) - len(str2)) / (len(str1) + len(str2))
        score = 80 - whitespace_ratio * 10
        return max(score, 0)
    levenshtein_ratio = fuzz.ratio(str1, str2)
    score = 85 * (levenshtein_ratio / 100)
    return max(score, 0)


ARGS = {
    "str1": "  hello world  ",
    "str2": "hello world"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
rapidfuzz — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 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 13 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

Large language models have been successfully applied to programming assistance tasks, such as code completion, code insertion, and instructional code editing. However, these applications remain insufficiently automated and struggle to effectively integrate various types of information during the programming process, including coding history, current code, and user instructions. In this work, we propose a new conversational framework that comprehensively integrates these information sources, collect data to train our models and evaluate their performance. Firstly, to thoroughly evaluate how well models align with different types of information and the quality of their outputs, we introduce a new benchmark, APEval (Assist Programming Eval), to comprehensively assess the performance of models in programming assistance tasks. Then, for data collection, we develop a data generation pipeline, Programming-Instruct, which synthesizes training data from diverse sources, such as GitHub and online judge platforms. This pipeline can automatically generate various types of messages throughout the programming process. Finally, using this pipeline, we generate 219K samples, fine-tune multiple models, and develop the CursorCore series. We show that CursorCore outperforms other models of comparable size. This framework unifies applications such as inline chat and automated editing, contributes to the advancement of coding assistants. Code, models and data are freely available at https://github.com/TechxGenus/CursorCore.

For agents

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

get_harvested_code_for_paper("2410.07002")
get_code_for_paper("2410.07002")
have("2410.07002")

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

curl https://syntology.ai/api/ran/2410.07002.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.07002.svg)](https://syntology.ai/paper/2410.07002)

Connect an agent — have() is free.