SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.06608 · 2024

The Prompt Report: A Systematic Survey of Prompt Engineering Techniques

arXiv · PDF · Open in the Atlas

Code that ran

3 of 5 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, 0 violated it and 0 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/2406.06608.json · badge: /api/ran/2406.06608.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
trigaten/Prompt_Systematic_Review canonical (classifier label)
classifier basis: census
3 ran · 2 no recorded run
FunctionStatusWhere it lives
evaluate_mmlu_response Ran
ran; our contract draft was wrong, not the code
trigaten/Prompt_Systematic_Review/src/prompt_systematic_review/experiments/benchmarking.py
code served (permissive licence) · get_code("7c810210b06d6018")
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
import json
from json.decoder import JSONDecodeError

class MockMessage:
    def __init__(self, content):
        self.content = content

class MockResponse:
    def __init__(self, message_content, is_json=False):
        self.message = MockMessage(message_content)

# Test case 1: JSON mode with correct answer
response_json_correct = MockResponse(
    '{"answer": "A"}',
    is_json=True
)

# Test case 2: Non-JSON mode with extracted answer
response_text = MockResponse(
    "The correct answer is (B). This option best explains the concept."
)

# Test case 3: Non-JSON mode with parentheses format
response_text_alt = MockResponse(
    "(C) is the most likely answer to this question."
)

ARGS = {
    "response": response_text,
    "correct_answer": "B",
    "json_mode": False
}
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
extract_predicted_answer Ran
ran; our contract draft was wrong, not the code
trigaten/Prompt_Systematic_Review/src/prompt_systematic_review/experiments/benchmarking.py
code served (permissive licence) · get_code("10114a988f110cf8")
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 find_parentheses_with_letters(text):
    """
    Finds letters A-D surrounded by parentheses.
    :param text: The text to search.
    :type text: str
    :return List[str]: The list of letters found.
    :rtype: List[str]
    """
    pattern = '\\(\\s*([A-D])\\s*\\)'
    matches = re.findall(pattern, text)
    return matches

def extract_predicted_answer(text):
    answer = None
    pattern = re.compile('(the correct answer is|the answer to the problem is|the answer is|the answer to the question is|the solution is|the right answer is|the correct solution is) \\(([A-D])\\)|\\(([A-D])\\) is (?:the correct answer|the right option|the most likely option)|the (?:correct answer is option|most likely answer is|most appropriate answer is|correct answer option is) \\(([A-D])\\)|\\(([A-D])\\) seems to be the most likely answer', re.IGNORECASE)
    match = pattern.search(text)
    if match:
        answer = next((match.group(i) for i in range(2, 7) if match.group(i) is not None))
    elif len(find_parentheses_with_letters(text)) == 1:
        answer = find_parentheses_with_letters(text)[0]
    return answer


# Setup code to construct valid arguments
text_with_answer = "The correct answer is (B) because it satisfies all conditions."

ARGS = {
    "text": text_with_answer
}
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
move_and_rename_extracted_contents Ran
ran; our contract draft was wrong, not the code
trigaten/Prompt_Systematic_Review/src/prompt_systematic_review/experiments/download_mmlu.py
code served (permissive licence) · get_code("6d756b13d340f97f")
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
import shutil
import tempfile

# Create temporary directories for testing
temp_base = tempfile.mkdtemp()
extracted_folder = tempfile.mkdtemp(dir=temp_base)
final_folder = tempfile.mkdtemp(dir=temp_base)

# Create some sample files/folders in the extracted_folder to move
sample_file_1 = os.path.join(extracted_folder, "file1.txt")
with open(sample_file_1, 'w') as f:
    f.write("sample content 1")

sample_file_2 = os.path.join(extracted_folder, "file2.txt")
with open(sample_file_2, 'w') as f:
    f.write("sample content 2")

sample_subdir = os.path.join(extracted_folder, "subdir")
os.makedirs(sample_subdir, exist_ok=True)
sample_file_3 = os.path.join(sample_subdir, "file3.txt")
with open(sample_file_3, 'w') as f:
    f.write("sample content 3")

new_folder_name = "mmlu_dataset"

ARGS = {
    "extracted_folder": extracted_folder,
    "final_folder": final_folder,
    "new_folder_name": new_folder_name
}
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
query_model Not yet run
unverified (harvested, no recorded run)
trigaten/Prompt_Systematic_Review/src/prompt_systematic_review/experiments/benchmarking.py
code served (permissive licence) · get_code("590036abca2df4b6")
How it ran

no invocation recorded.

review_abstract_title_categorical Not yet run
unverified (harvested, no recorded run)
trigaten/Prompt_Systematic_Review/src/prompt_systematic_review/automated_review.py
code served (permissive licence) · get_code("af498d3aa50682eb")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

Generative Artificial Intelligence (GenAI) systems are increasingly being deployed across diverse industries and research domains. Developers and end-users interact with these systems through the use of prompting and prompt engineering. Although prompt engineering is a widely adopted and extensively researched area, it suffers from conflicting terminology and a fragmented ontological understanding of what constitutes an effective prompt due to its relatively recent emergence. We establish a structured understanding of prompt engineering by assembling a taxonomy of prompting techniques and analyzing their applications. We present a detailed vocabulary of 33 vocabulary terms, a taxonomy of 58 LLM prompting techniques, and 40 techniques for other modalities. Additionally, we provide best practices and guidelines for prompt engineering, including advice for prompting state-of-the-art (SOTA) LLMs such as ChatGPT. We further present a meta-analysis of the entire literature on natural language prefix-prompting. As a culmination of these efforts, this paper presents the most comprehensive survey on prompt engineering to date.

For agents

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

get_harvested_code_for_paper("2406.06608")
get_code_for_paper("2406.06608")
have("2406.06608")

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

curl https://syntology.ai/api/ran/2406.06608.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.06608.svg)](https://syntology.ai/paper/2406.06608)

Connect an agent — have() is free.