bioguider 0.2.12__tar.gz → 0.2.14__tar.gz

{bioguider-0.2.12 → bioguider-0.2.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: bioguider
-Version: 0.2.12
+Version: 0.2.14
 Summary: An AI-Powered package to help biomedical developers to generate clear documentation
 License: MIT
 Author: Cankun Wang

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/agent_task.py

@@ -13,7 +13,12 @@ class AgentTask(ABC):
     A class representing a step in an agent's process.
     """
 
-    def __init__(self, llm: BaseChatOpenAI, step_callback: Callable | None = None):
+    def __init__(
+        self, 
+        llm: BaseChatOpenAI, 
+        step_callback: Callable | None = None,
+        summarized_files_db: SummarizedFilesDb | None = None,
+    ):
         """
         Initialize the AgentStep with a language model and a callback function.
 
@@ -23,7 +28,7 @@ class AgentTask(ABC):
         """
         self.llm = llm
         self.step_callback = step_callback
-        self.summary_file_db = None
+        self.summarized_files_db = summarized_files_db
         self.graph: CompiledGraph | None = None
 
     def _print_step(
@@ -45,7 +50,7 @@ class AgentTask(ABC):
             token_usage=token_usage,
         )
 
-    def compile(self, repo_path: str, gitignore_path: str, db: SummarizedFilesDb | None = None, **kwargs):
+    def compile(self, repo_path: str, gitignore_path: str, **kwargs):
         """
         Compile the agent step with the given repository and gitignore paths.
 
@@ -55,7 +60,6 @@ class AgentTask(ABC):
             **kwargs: derived class may pass more arguments to implmented _compile(), that is,
                 what **kwargs is depends on derived class
         """
-        self.summary_file_db = db
         self._compile(repo_path, gitignore_path, **kwargs)
     
     @abstractmethod

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/agent_tools.py

@@ -1,4 +1,5 @@
 import os
+import logging
 from typing import Callable
 from markdownify import markdownify as md
 from langchain_openai.chat_models.base import BaseChatOpenAI
@@ -7,6 +8,8 @@ from bioguider.utils.file_utils import get_file_type
 from bioguider.agents.agent_utils import read_directory, read_file, summarize_file
 from bioguider.rag.data_pipeline import count_tokens
 
+logger = logging.getLogger(__name__)
+
 class agent_tool:
     def __init__(
         self,
@@ -53,19 +56,12 @@ Returns:
 class summarize_file_tool(agent_tool):
     """ Read a file and generate a summary according to a specified prompt.
 
-Arguments
-----------
-    file_path : str, required
-        Path to the file to read.
-    summarize_prompt : str, optional
-        Instruction guiding the summarization focus (default is "N/A").
-        Use this to emphasize specific aspects of the content.
+Args:
+    file_path str: required. The file path to read.
+    summarize_prompt str: optional. A string instruction guiding the summarization focus (default is "N/A"). Use this to emphasize specific aspects of the content.
 
-Returns
--------
-    str or None
-        A summarized version of the file content.
-        Returns None if the file does not exist or cannot be read.     
+Returns:
+    str or None: A summarized version of the file content. Returns None if the file does not exist or cannot be read.     
     """
     def __init__(
         self, 
@@ -124,8 +120,15 @@ Returns
         if summarized_content is not None:
             return f"summarized content of file {file_path}: " + summarized_content
 
-        file_content = read_file(abs_file_path)
-        file_content = file_content.replace("{", "{{").replace("}", "}}")
+        try:
+            file_content = read_file(abs_file_path)
+            file_content = file_content.replace("{", "{{").replace("}", "}}")
+        except UnicodeDecodeError as e:
+            logger.error(str(e))
+            return f"{file_path} is a binary, can't be summarized."
+        except Exception as e:
+            logger.error(str(e))
+            return f"Failed to read {file_path}."
         summarized_content, token_usage = summarize_file(
             self.llm, abs_file_path, file_content, self.detailed_level,
             summary_instructions=self.summarize_instruction,

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/agent_utils.py

@@ -16,11 +16,12 @@ from langchain.tools import BaseTool
 from langchain.schema import AgentAction, AgentFinish
 from langchain.agents import AgentOutputParser
 from langgraph.prebuilt import create_react_agent
+from langchain_community.callbacks.openai_info import OpenAICallbackHandler
 import logging
 
 from pydantic import BaseModel, Field
 
-from bioguider.utils.constants import DEFAULT_TOKEN_USAGE
+from bioguider.utils.constants import DEFAULT_TOKEN_USAGE, MAX_FILE_LENGTH, MAX_SENTENCE_NUM
 from bioguider.utils.file_utils import get_file_type
 from ..utils.gitignore_checker import GitignoreChecker
 from ..database.summarized_file_db import SummarizedFilesDb
@@ -178,8 +179,7 @@ Here is the file content:
 Now, let's start to summarize.
 """)
 
-MAX_FILE_LENGTH=20 *1024 # 20K
-MAX_SENTENCE_NUM=20
+
 def summarize_file(
     llm: BaseChatOpenAI, 
     name: str, 
@@ -379,6 +379,20 @@ def escape_braces(text: str) -> str:
     text = re.sub(r'(?<!{){(?!{)', '{{', text)
     return text
 
+STRING_TO_OBJECT_SYSTEM_PROMPT = """
+You are an expert to understand data. You will be provided a text, and your task is to extracted structured data from the provided text.
+
+---
+
+### **Instructions**
+1. If no structured data can be extracted, return None
+
+---
+
+### **Input Text**
+{input_text}
+"""
+
 def try_parse_json_object(json_obj: str) -> dict | None:
     json_obj = json_obj.strip()
 
@@ -406,4 +420,26 @@ def try_parse_json_object(json_obj: str) -> dict | None:
         return None
     except Exception as e:
         logger.error(e)
-        return None
+        return None
+    
+def try_parse_with_llm(llm: BaseChatOpenAI, input_text: str, schema: any):
+    system_prompt = ChatPromptTemplate.from_template(
+        STRING_TO_OBJECT_SYSTEM_PROMPT
+    ).format(input_text=input_text)
+    prompt = ChatPromptTemplate.from_messages([
+        ("system", system_prompt)
+    ])
+    agent = prompt | llm.with_structured_output(schema)
+    callback_handler = OpenAICallbackHandler()
+
+    try:
+        res = agent.invoke(
+            input={},
+            config={
+                "callbacks": [callback_handler],
+            },
+        )
+        return res, vars(callback_handler)
+    except Exception as e:
+        logger.error(e)
+        return None

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/collection_observe_step.py

@@ -5,7 +5,7 @@ from langchain_openai.chat_models.base import BaseChatOpenAI
 from langchain_core.prompts import ChatPromptTemplate
 from bioguider.agents.agent_utils import ObservationResult
 from bioguider.agents.collection_task_utils import CollectionWorkflowState
-from bioguider.agents.common_agent_2step import CommonAgentTwoSteps
+from bioguider.agents.common_agent_2step import CommonAgentTwoChainSteps, CommonAgentTwoSteps
 from bioguider.agents.peo_common_step import PEOCommonStep
 from bioguider.agents.prompt_utils import COLLECTION_GOAL, COLLECTION_PROMPTS
 
@@ -34,11 +34,13 @@ Here is the 2-level file structure of the repository (`f` = file, `d` = director
 
   * Provide your reasoning under **Analysis**
   * Then list all relevant files and folders under **FinalAnswer**
+  * **FinalAnswer** format must exactly match this format:
+    **FinalAnswer**: {{"final_answer": [<file path>, <file path>, <file path>, ...]}}
   * Be sure to include the **full relative paths** with respect to the repository root.
-  * Your answer **must follow this exact format** (note: no JSON code block, no additional comments):
+  * Your answer **must exactly match the follwing format** (note: no JSON code block, no additional comments), **do not** make up anything:
 
   ```
-  **Analysis**: your analysis here  
+  **Analysis**: your analysis here 
   **FinalAnswer**: {{"final_answer": ["path/to/file1", "path/to/file2", ...]}}
   ```
 4. If you believe **more files still need to be collected**:
@@ -80,8 +82,8 @@ class CollectionObserveStep(PEOCommonStep):
         repo_structure = self.repo_structure
         intermediate_steps = self._build_intermediate_steps(state)
         prompt = ChatPromptTemplate.from_template(COLLECTION_OBSERVE_SYSTEM_PROMPT)
-        important_instructions = "N/A" if "important_instructions" not in collection_item or len(collection_item["important_instructions"]) == 0 \
-            else collection_item["important_instructions"]
+        important_instructions = "N/A" if "observe_important_instructions" not in collection_item or len(collection_item["observe_important_instructions"]) == 0 \
+            else collection_item["observe_important_instructions"]
         return prompt.format(
             goal_item_desc=goal_item_desc,
             related_file_description=collection_item["related_file_description"],

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/collection_plan_step.py

@@ -8,7 +8,7 @@ from bioguider.agents.agent_utils import (
     PlanAgentResultJsonSchema,
     PlanAgentResult,
 )
-from bioguider.agents.common_agent_2step import CommonAgentTwoSteps
+from bioguider.agents.common_agent_2step import CommonAgentTwoChainSteps, CommonAgentTwoSteps
 from bioguider.agents.peo_common_step import PEOCommonStep
 from bioguider.agents.collection_task_utils import CollectionWorkflowState
 from bioguider.agents.prompt_utils import COLLECTION_GOAL, COLLECTION_PROMPTS
@@ -57,7 +57,9 @@ Here are the results from previous steps:
 
 3. You may use the `read_directory` tool to explore directory contents, but avoid using it in the first step unless necessary.
 
-4. You may use the `python_repl` tool to execute Python code, but this should **also be avoided in the first step**.
+4. Your plan can only use the above tools, **do not** make up any tools not in the above tools list.
+
+5. Your planned step input file or input directory must come from the above repository files structure, **do not** make up file name or directory name.
 
 ---
 
@@ -65,12 +67,12 @@ Here are the results from previous steps:
 {important_instructions}
 
 ### **Output Format**  
-Your plan should be returned as a sequence of steps in the following format:
+Your plan **must exactly match** a sequence of steps in the following format, **do not** make up anything:
 
-Step: <tool name>   # Tool name must be one of {tool_names}  
+Step: <tool name>   # Tool name **must be one** of {tool_names}  
 Step Input: <file or directory name>
 
-Step: <tool name>  
+Step: <tool name>  # Tool name **must be one** of {tool_names}  
 Step Input: <file or directory name>
 ...
 """)
@@ -105,8 +107,8 @@ class CollectionPlanStep(PEOCommonStep):
         step_analysis, step_thoughts = self._build_intermediate_analysis_and_thoughts(state)
         goal = ChatPromptTemplate.from_template(COLLECTION_GOAL).format(goal_item=collection_item["goal_item"])
         related_file_description = collection_item["related_file_description"]
-        important_instructions="N/A" if "important_instructions" not in collection_item or len(collection_item["important_instructions"]) == 0 \
-            else collection_item["important_instructions"]
+        important_instructions="N/A" if "plan_important_instructions" not in collection_item or len(collection_item["plan_important_instructions"]) == 0 \
+            else collection_item["plan_important_instructions"]
         tool_names, tools_desc = get_tool_names_and_descriptions(self.custom_tools)
         system_prompt = COLLECTION_PLAN_SYSTEM_PROMPT.format(
             goal=goal,

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/collection_task.py

@@ -50,9 +50,12 @@ class CollectionTask(AgentTask):
     def __init__(
         self, 
         llm: BaseChatOpenAI, 
-        step_callback: Callable | None = None
+        step_callback: Callable | None = None,
+        summarize_instruction: str | None = "N/A",
+        summarized_files_db: SummarizedFilesDb | None = None,
+        provided_files: list[str] | None = None,
     ):
-        super().__init__(llm, step_callback)
+        super().__init__(llm, step_callback, summarized_files_db=summarized_files_db)
         self.repo_path: str | None = None
         self.gitignore_path: str | None = None
         self.repo_structure: str | None = None
@@ -60,6 +63,8 @@ class CollectionTask(AgentTask):
         self.steps: list[PEOCommonStep] = []
         self.tools: list[any] | None = None
         self.custom_tools: list[Tool] | None = None
+        self.summarize_instruction = summarize_instruction
+        self.provided_files = provided_files
 
     def _prepare_tools(self, related_file_goal_item_desc):
         tool_rd = read_directory_tool(repo_path=self.repo_path)
@@ -67,7 +72,8 @@ class CollectionTask(AgentTask):
             llm=self.llm,
             repo_path=self.repo_path,
             output_callback=self.step_callback,
-            db=self.summary_file_db,
+            db=self.summarized_files_db,
+            summaize_instruction=self.summarize_instruction,
         )
         tool_rf = read_file_tool(repo_path=self.repo_path)
         tool_cf = check_file_related_tool(
@@ -75,6 +81,8 @@ class CollectionTask(AgentTask):
             repo_path=self.repo_path,
             goal_item_desc=related_file_goal_item_desc,
             output_callback=self.step_callback,
+            summarize_instruction=self.summarize_instruction,
+            summarized_files_db=self.summarized_files_db,
         )
         self.tools = [tool_rd, tool_sum, tool_rf, tool_cf]
         self.custom_tools = [
@@ -99,13 +107,15 @@ class CollectionTask(AgentTask):
                 description=tool_cf.__class__.__doc__,
             ),
         ]
-        self.custom_tools.append(CustomPythonAstREPLTool())
+        # self.custom_tools.append(CustomPythonAstREPLTool())
 
     def _initialize(self):
         # initialize the 2-level file structure of the repo
         if not os.path.exists(self.repo_path):
             raise ValueError(f"Repository path {self.repo_path} does not exist.")
-        files = read_directory(self.repo_path, os.path.join(self.repo_path, ".gitignore"))
+        files = self.provided_files
+        if files is None:
+            files = read_directory(self.repo_path, os.path.join(self.repo_path, ".gitignore"))
         file_pairs = [(f, get_file_type(os.path.join(self.repo_path, f)).value) for f in files]
         self.repo_structure = ""
         for f, f_type in file_pairs:

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/collection_task_utils.py

@@ -4,13 +4,17 @@ from langchain.prompts import ChatPromptTemplate
 from langchain_openai.chat_models.base import BaseChatOpenAI
 from langchain_core.messages import AIMessage
 from pydantic import BaseModel, Field
+import logging
 
 from bioguider.agents.agent_tools import agent_tool
 from bioguider.agents.agent_utils import read_file, summarize_file
 from bioguider.agents.peo_common_step import PEOWorkflowState
 from bioguider.agents.common_agent import CommonAgent
 from bioguider.agents.common_agent_2step import CommonAgentTwoSteps
+from bioguider.database.summarized_file_db import SummarizedFilesDb
+from bioguider.utils.constants import MAX_FILE_LENGTH
 
+logger = logging.getLogger(__name__)
 
 class CollectionWorkflowState(TypedDict):
     llm: Optional[BaseChatOpenAI]
@@ -46,20 +50,22 @@ Does this file appear to contain related information?
 
 ---
 
-### **Output Format:**  
-Respond with a single word: "Yes" or "No" to indicate whether the file is related to the goal item.
-Do not include any additional text, explanation, or formatting.
+### **Output Format:** 
+Respond with exactly two parts:
+1. A single word: Yes or No (indicating if the file meets the goal criteria)
+2. One brief explanatory sentence.
+For example: Yes. This file is a compiled binary file, so, it is related to the compiled standalone file (goal item).
 """)
 
 class CheckFileRelatedResult(BaseModel):
-    is_related: bool = Field(description="True if the file is related to the goal item, False otherwise.")
+    is_related: str = Field(description="A string conclusion specify if the provided file is related. The string value contains two parts:\n 1. A single word: Yes or No (indicating if the file meets the goal criteria).\n 2. One brief explanatory sentence.")
 
 class check_file_related_tool(agent_tool):
     """ Check if the file is related to the goal item
 Args:
     file_path str: file path
 Returns:
-    bool: True if the file is related to the goal item, False otherwise.
+    str: A string conclusion. The string conclusion contains two parts:\n 1. A single word: Yes or No (indicating if the file meets the goal criteria).\n 2. One brief explanatory sentence.
     """ 
     def __init__(
         self, 
@@ -67,23 +73,51 @@ Returns:
         repo_path: str,
         goal_item_desc: str,
         output_callback: Callable | None = None,
+        summarize_instruction: str | None = None,
+        summarize_level: int | None = 6,
+        summarized_files_db: SummarizedFilesDb | None = None,
     ):
         super().__init__(llm=llm, output_callback=output_callback)
         self.repo_path = repo_path
         self.goal_item_desc = goal_item_desc
+        self.summarize_instruction = summarize_instruction \
+            if summarize_instruction is not None else "N/A"
+        self.summarize_level = summarize_level
+        self.summarized_files_db = summarized_files_db
 
     def run(self, file_path: str) -> str:
         if not self.repo_path in file_path:
             file_path = os.path.join(self.repo_path, file_path)
         if not os.path.isfile(file_path):
             return "Can't read file"
-        file_content = read_file(file_path)
-        if file_content is None:
+        
+        check_prompts = None
+        try:
+            file_content = read_file(file_path)
+        except UnicodeDecodeError as e:
+            logger.error(str(e))
+            check_prompts = "Can't summarize binary file, please decide according to file name and extension."
+        except Exception as e:
+            logger.error(str(e))
+            check_prompts = "Failed to summarize file, please decide according to file name and extension."
+        if check_prompts is None and file_content is None:
             return "Failed to read file"
-        summarized_content, token_usage = summarize_file(self.llm, file_path, file_content, 6)
-        if summarized_content is None:
-            return "Failed to summarize file"
-        self._print_token_usage(token_usage)
+        if check_prompts is not None:
+            summarized_content = check_prompts
+        else:
+            if len(file_content) > MAX_FILE_LENGTH:
+                file_content = file_content[:MAX_FILE_LENGTH]
+            summarized_content, token_usage = summarize_file(
+                llm=self.llm, 
+                name=file_path, 
+                content=file_content, 
+                level=self.summarize_level,
+                summary_instructions=self.summarize_instruction,
+                db=self.summarized_files_db,
+            )
+            if summarized_content is None:
+                return "Failed to summarize file"
+            self._print_token_usage(token_usage)
         
         prompt = CHECK_FILE_RELATED_USER_PROMPT.format(
             goal_item_desc=self.goal_item_desc,
@@ -102,8 +136,5 @@ Returns:
         
         self._print_step_output(step_output=reasoning)
         self._print_token_usage(token_usage)
-        if out:
-            return "Yes, the file is related to the goal item."
-        else:
-            return "No, the file **is not** related to the goal item."
+        return res.is_related
         

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/dockergeneration_task.py

@@ -47,7 +47,7 @@ class DockerGenerationTask(AgentTask):
     def __init__(
         self, 
         llm, 
-        step_callback = None
+        step_callback = None,
     ):
         super().__init__(llm, step_callback)
         self.repo_path: str | None = None

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/evaluation_installation_task.py

@@ -9,7 +9,8 @@ from pydantic import BaseModel, Field
 from markdownify import markdownify as md
 
 from bioguider.agents.agent_utils import read_file
-from bioguider.agents.prompt_utils import EVALUATION_INSTRUCTION
+from bioguider.agents.collection_task import CollectionTask
+from bioguider.agents.prompt_utils import EVALUATION_INSTRUCTION, CollectionGoalItemEnum
 from bioguider.utils.constants import DEFAULT_TOKEN_USAGE, ProjectMetadata
 from bioguider.rag.data_pipeline import count_tokens
 from .common_agent_2step import CommonAgentTwoSteps, CommonAgentTwoChainSteps
@@ -32,14 +33,17 @@ Your task is to analyze the provided files related to installation and generate
 1. **Installation Available**: Is the installation section in document (like README.md or INSTALLATION)?
    * Output: `Yes` or `No`
 
-2. **Installation Tutorial**: Is the installation tutorial provided?
+2. **Installation Tutorial**: Is the step-by-step installation tutorial provided?
    * Ouput: `Yes` or `No`
 
 3. **Number of required Dependencies Installation**: The number of dependencies that are required to install
    * Output: Number
    * Suggest specific improvements if necessary, such as missing dependencies
 
-4. **Overall Score**: Give an overall quality rating of the Installation information.
+4. **Compatible Operating System**: Is the compatible operating system described?
+   * Output: `Yes` or `No`
+
+5. **Overall Score**: Give an overall quality rating of the Installation information.
    * Output: `Poor`, `Fair`, `Good`, or `Excellent`
 
 ---
@@ -53,6 +57,7 @@ Your final report must **exactly match** the following format. Do not add or omi
 **Dependency:**
   * number: [Number]
   * suggestions: <suggestion to improve **dependency information** like missing dependencies
+**Compatible Operating System:** [Yes / No]
 **Overall Score:** [Poor / Fair / Good / Excellent]
 
 ---
@@ -113,6 +118,7 @@ class StructuredEvaluationInstallationResult(BaseModel):
     install_tutorial: Optional[bool]=Field(description="A boolean value. Is the installation tutorial provided?")
     dependency_number: Optional[int]=Field(description="A number. It is the number of dependencies that are required to install.")
     dependency_suggestions: Optional[str]=Field(description="A string value. It is the specific improvements if necessary, such as missing dependencies")
+    compatible_os: Optional[bool]=Field(description="A boolean value. Is compatible operating system described?")
     overall_score: Optional[str]=Field(description="A overall scroll for the installation quality, could be `Poor`, `Fair`, `Good`, or `Excellent`")
 
 class EvaluationInstallationResult(BaseModel):
@@ -163,8 +169,9 @@ class EvaluationInstallationTask(EvaluationTask):
         gitignore_path, 
         meta_data = None, 
         step_callback = None,
+        summarized_files_db = None,
     ):
-        super().__init__(llm, repo_path, gitignore_path, meta_data, step_callback)
+        super().__init__(llm, repo_path, gitignore_path, meta_data, step_callback, summarized_files_db)
         self.evaluation_name = "Installation Evaluation"
 
 
@@ -204,6 +211,8 @@ class EvaluationInstallationTask(EvaluationTask):
             instruction_prompt=EVALUATION_INSTRUCTION,
             schema=StructuredEvaluationInstallationResult,
         )
+        res: StructuredEvaluationInstallationResult = res
+        res.dependency_number = 0 if res.dependency_number is None else res.dependency_number
         self.print_step(step_output=reasoning_process)
         self.print_step(token_usage=token_usage)
 
@@ -235,7 +244,7 @@ class EvaluationInstallationTask(EvaluationTask):
         }
         return evaluation, token_usage
     
-    def _evaluate(self, files: list[str] | None = None) -> tuple[dict | None, dict]:
+    def _evaluate(self, files: list[str] | None = None) -> tuple[dict | None, dict, list[str]]:
         evaluation, token_usage = self._free_evaluate(files)
         structured_evaluation, structured_token_usage = self._structured_evaluate(files)
 
@@ -245,5 +254,20 @@ class EvaluationInstallationTask(EvaluationTask):
         }
         total_token_usage = increase_token_usage(token_usage, structured_token_usage)
 
-        return combined_evaluation, total_token_usage
-        
+        return combined_evaluation, total_token_usage, files
+
+    def _collect_files(self):
+        task = CollectionTask(
+            llm=self.llm,
+            step_callback=self.step_callback,
+        )
+        task.compile(
+            repo_path=self.repo_path,
+            gitignore_path=Path(self.repo_path, ".gitignore"),
+            db=self.summarized_files_db,
+            goal_item=CollectionGoalItemEnum.Installation.name,
+        )
+        files = task.collect()
+        if files is None:
+            return []
+        return files

{bioguider-0.2.12 → bioguider-0.2.14}/bioguider/agents/evaluation_readme_task.py

@@ -7,6 +7,7 @@ from langchain_openai.chat_models.base import BaseChatOpenAI
 from pydantic import BaseModel, Field
 
 from bioguider.agents.prompt_utils import EVALUATION_INSTRUCTION
+from bioguider.utils.gitignore_checker import GitignoreChecker
 
 from ..utils.pyphen_utils import PyphenReadability
 from bioguider.agents.agent_utils import increase_token_usage, read_file, summarize_file
@@ -303,9 +304,10 @@ class EvaluationREADMETask(EvaluationTask):
         repo_path: str, 
         gitignore_path: str,
         meta_data: ProjectMetadata | None = None,
-        step_callback: Callable | None = None
+        step_callback: Callable | None = None,
+        summarized_files_db = None,
     ):
-        super().__init__(llm, repo_path, gitignore_path, meta_data, step_callback)
+        super().__init__(llm, repo_path, gitignore_path, meta_data, step_callback, summarized_files_db)
         self.evaluation_name = "README Evaluation"
 
     def _structured_evaluate(self, free_readme_evaluations: dict[str, dict]):
@@ -455,7 +457,7 @@ class EvaluationREADMETask(EvaluationTask):
             total_token_usage = increase_token_usage(total_token_usage, token_usage)
         return readme_evaluations, total_token_usage
             
-    def _evaluate(self, files: list[str]) -> tuple[dict, dict]:
+    def _evaluate(self, files: list[str]) -> tuple[dict, dict, list[str]]:
         free_readme_evaluations, free_token_usage = self._free_evaluate(files)
         structured_readme_evaluations, structured_token_usage = self._structured_evaluate(free_readme_evaluations)
 
@@ -472,6 +474,26 @@ class EvaluationREADMETask(EvaluationTask):
         
         total_token_usage = increase_token_usage(free_token_usage, structured_token_usage)
 
-        return combined_evaluations, total_token_usage
+        return combined_evaluations, total_token_usage, files
     
+    def _collect_files(self):
+        """
+        Search for a README file in the repository directory.
+        """
+        possible_readme_files = [
+            "readme.md",
+            "readme.rst",
+            "readme.txt",
+            "readme",
+        ]
+        repo_path = self.repo_path
+        gitignore_path = Path(repo_path, ".gitignore")
+        gitignore_checker = GitignoreChecker(
+            directory=repo_path, gitignore_path=gitignore_path
+        )
+        found_readme_files = gitignore_checker.check_files_and_folders(
+            check_file_cb=lambda root_dir, relative_path: Path(relative_path).name.lower() in possible_readme_files,
+        )
+                
+        return found_readme_files