Clean up code

agent/agent_config/prompts.py

@@ -29,7 +29,4 @@ system_message = {
         "Do not attempt to use any other tools such as web_search."
         "DO NOT HALLUCINATE OR MAKE UP TOOLS."
     )
-}
-
-#         "STRICTLY use the `retrieve_context` tool to get the relevant code snippets or metadata about the codebase to formulate your response.\n\n"
-        # "Stick to the context that your retri"
+}

agent/core.py

@@ -61,7 +61,7 @@ async def run_agent(issue_url: str, branch_name: str = "main") -> str:
                         issue_title = function_result.get("title")
                         issue_body = function_result.get("body")
                         issue_description_cache = issue_title + "\n" + issue_body if issue_title or issue_body else None
-                        print("ISSUE DESCRIPTION CACHE ✨:", issue_description_cache)
+                        print("ISSUE DESCRIPTION CACHE:", issue_description_cache)
 
                     if function_name == "retrieve_context":
                         if "issue_description" in function_params:
@@ -69,7 +69,7 @@ async def run_agent(issue_url: str, branch_name: str = "main") -> str:
                                 issue_description_cache
                                 and (function_params["issue_description"] != issue_description_cache)
                             ):
-                                print("🔁 Overriding incorrect issue_description with correct one from cache.")
+                                print("Overriding incorrect issue_description with correct one from cache.")
                                 function_params["issue_description"] = issue_description_cache
                                 function_result = names_to_functions[function_name](**function_params)
 

tools/code_index.py

@@ -2,14 +2,12 @@ import asyncio
 import numpy as np
 import os
 from sklearn.metrics.pairwise import cosine_similarity
-import time
-from typing import List, Dict
+from typing import List
 from llama_index.core import VectorStoreIndex, Document, Settings, get_response_synthesizer
 from llama_index.core.query_engine import RetrieverQueryEngine
 from llama_index.core.postprocessor import SimilarityPostprocessor
 from llama_index.embeddings.mistralai import MistralAIEmbedding
 from llama_index.llms.mistralai import MistralAI
-from mistralai import Mistral
 from config import MISTRAL_API_KEY
 from tools.utils import fetch_repo_files, fetch_file_content
 
@@ -61,25 +59,6 @@ def select_relevant_files_semantic(issue_description: str, file_paths: List[str]
 
     return top_files
 
-
-# print(select_relevant_files_semantic(
-# '''
-# 🛠️ Configuration Error: Placeholder values detected in host_config.json
-# This file still includes default placeholders like:
-
-# <evalai_user_auth_token>
-# <host_team_pk>
-# <evalai_host_url>
-# Please replace them with real values to proceed.
-# ''',
-# ['.github/FUNDING.yml', '.github/workflows/process_challenge.yml', '.gitignore', 'README.md', 'annotations/test_annotations_devsplit.json', 'annotations/test_annotations_testsplit.json', 'challenge_config.yaml', 'challenge_data/__init__.py', 'challenge_data/challenge_1/__init__.py', 'challenge_data/challenge_1/main.py', 'evaluation_script/__init__.py', 'evaluation_script/main.py', 'github/challenge_processing_script.py', 'github/config.py', 'github/host_config.json', 'github/requirements.txt', 'github/utils.py', 'logo.jpg', 'remote_challenge_evaluation/README.md', 'remote_challenge_evaluation/eval_ai_interface.py', 'remote_challenge_evaluation/evaluate.py', 'remote_challenge_evaluation/main.py', 'remote_challenge_evaluation/requirements.txt', 'run.sh', 'submission.json', 'templates/challenge_phase_1_description.html', 'templates/challenge_phase_2_description.html', 'templates/description.html', 'templates/evaluation_details.html', 'templates/submission_guidelines.html', 'templates/terms_and_conditions.html', 'worker/__init__.py', 'worker/run.py']))
-
-
-# Assuming these are async now or wrapped appropriately
-# async def fetch_repo_files(...)
-# async def fetch_file_content(...)
-# async def VectorStoreIndex.from_documents(...)
-
 async def async_retry_on_429(func, *args, max_retries=3, delay=1, **kwargs):
     for attempt in range(max_retries):
         try:
@@ -128,20 +107,7 @@ async def build_repo_index(owner: str, repo: str, ref: str = "main", issue_descr
     return index
 
 
-# print(build_repo_index("aditi-dsi", "EvalAI-Starters", "master", 
-    # '''
-    # 🛠️ Configuration Error: Placeholder values detected in host_config.json
-    # This file still includes default placeholders like:
-
-    # <evalai_user_auth_token>
-    # <host_team_pk>
-    # <evalai_host_url>
-    # Please replace them with real values to proceed.
-    # '''))
-
-
 async def retrieve_context(owner: str, repo: str, ref: str, issue_description: str) -> List[str]:
-    print("Issue Description:", issue_description)
     index = await build_repo_index(owner, repo, ref, issue_description)
     Settings.llm = MistralAI(model="codestral-latest", api_key=MISTRAL_API_KEY)
     Settings.embed_model = MistralAIEmbedding(model_name="codestral-embed", api_key=MISTRAL_API_KEY)
@@ -165,21 +131,7 @@ async def retrieve_context(owner: str, repo: str, ref: str, issue_description: s
         "- DO NOT include generic, loosely related, or unrelated content.\n"
     )
 
-    print("Query:", query)
-
-    # If query_engine.query is sync, wrap it in a thread
     response = await asyncio.to_thread(query_engine.query, query)
 
     print(response)
-    return response
-
-# print(retrieve_context("aditi-dsi", "EvalAI-Starters", "master",
-#     '''
-#     🛠️ Configuration Error: Placeholder values detected in host_config.json
-#     This file still includes default placeholders like:
-
-#     <evalai_user_auth_token>
-#     <host_team_pk>
-#     <evalai_host_url>
-#     Please replace them with real values to proceed.
-#     '''))
+    return response

tools/github_tools.py

@@ -27,7 +27,6 @@ def get_issue_details(owner, repo, issue_num):
     else:
         raise Exception(f"Failed to fetch issue: {response.status_code} {response.text}")
 
-# print(get_issue_details("aditi-dsi", "testing-cryptope", "4"))
 
 def post_comment(owner, repo, issue_num, comment_body):
     installation_id = get_installation_id(owner, repo)
@@ -43,5 +42,3 @@ def post_comment(owner, repo, issue_num, comment_body):
         return response.json()
     else:
         raise Exception(f"Failed to post comment: {response.status_code} {response.text}")
-
-# print(post_comment("aditi-dsi", "testing-cryptope", "3", "This is a test comment from OpenSorus."))

tools/utils.py

@@ -70,8 +70,6 @@ def get_installation_id(owner, repo):
     else:
         raise Exception(f"Failed to get installation ID for {owner}/{repo}: {response.status_code} {response.text}")
     
-# print(get_installation_id("aditi-dsi", "testing-cryptope"))
-
 
 def get_installation_token(installation_id):
     """Return a valid installation token, fetch new if expired or missing."""
@@ -92,7 +90,6 @@ def get_installation_token(installation_id):
         installation_tokens[installation_id] = {"token": token, "expires_at": expires_at}
         return token
 
-# print(get_installation_token(69452220))
 
 async def fetch_repo_files(owner: str, repo: str, ref: str = "main") -> List[str]:
     """
@@ -115,7 +112,6 @@ async def fetch_repo_files(owner: str, repo: str, ref: str = "main") -> List[str
     file_paths = [item["path"] for item in tree if item["type"] == "blob"]
     return file_paths
 
-# print(fetch_repo_files("aditi-dsi", "EvalAI-Starters", "master"))
 
 async def fetch_file_content(owner: str, repo: str, path: str, ref: str = "main") -> str:
     """
@@ -137,5 +133,3 @@ async def fetch_file_content(owner: str, repo: str, path: str, ref: str = "main"
     content_json = response.json()
     content = base64.b64decode(content_json["content"]).decode("utf-8", errors="ignore")
     return content
-
-# print(fetch_file_content("aditi-dsi", "testing-cryptope", "frontend/src/lib/buildSwap.ts", "main"))