Fixed #62 : unify AI agents under FastMCP and implement hybrid RAG

MCP_Enhancement/agent_start.py

@@ -0,0 +1,127 @@
+from fastmcp import FastMCP
+import os
+from dotenv import load_dotenv
+from atlassian import Jira
+from github import Github
+
+# Initialize FastMCP Server
+mcp = FastMCP("Mifos Knowledge Librarian")
+
+# Load credentials
+load_dotenv()
+
+# Initialize clients
+jira_client = Jira(
+    url=os.environ["JIRA_INSTANCE_URL"],
+    username=os.environ["JIRA_USERNAME"],
+    password=os.environ["JIRA_API_TOKEN"]
+)
+github_client = Github(os.environ["GITHUB_TOKEN"])
+
+@mcp.tool()
+def search_jira_tickets(query: str):
+    """
+    Search for Jira tickets using JQL.
+    Used for tracking feature progress and identifying developer consensus.
+    """
+    try:
+        issues = jira_client.jql(query, limit=5)
+        results = []
+        for issue_data in issues['issues']:
+            # The jql result gives us the key and summary, let's get comments
+            comments_data = jira_client.get_issue_comments(issue_data['key'])
+            comments = [
+                f"{c['author']['displayName']}: {c['body']}"
+                for c in comments_data['comments'][-3:]
+            ]
+            results.append({
+                "key": issue_data['key'],
+                "summary": issue_data['fields']['summary'],
+                "comments": comments,
+            })
+        return results
+    except Exception as e:
+        return f"Error searching Jira: {e}"
+
+@mcp.tool()
+def get_github_pr_details(pr_number: int, repo_name: str):
+    """
+    Fetch description and changed files from a specific GitHub Pull Request.
+    Used for synthesizing release notes.
+    """
+    try:
+        repo = github_client.get_repo(repo_name)
+        pr = repo.get_pull(pr_number)
+        files = [f.filename for f in pr.get_files()]
+        return {
+            "title": pr.title,
+            "description": pr.body,
+            "changed_files": files,
+        }
+    except Exception as e:
+        return f"Error fetching GitHub PR: {e}"
+
+@mcp.tool()
+def generate_knowledge_summary(jira_key: str, repo_name: str = "apache/fineract"):
+    """
+    Takes a Jira Key, searches for a corresponding GitHub PR in a specified repo,
+    and returns a combined summary of the ticket and the PR.
+    Defaults to the 'apache/fineract' repository.
+    """
+    try:
+        # 1. Get Jira ticket info
+        jira_info_list = search_jira_tickets(f'key = {jira_key}')
+        if not jira_info_list or isinstance(jira_info_list, str):
+            return f"Could not retrieve Jira ticket {jira_key}."
+        jira_info = jira_info_list[0]
+
+        # 2. Find corresponding GitHub PR
+        query = f'repo:{repo_name} is:pr "{jira_key}"'
+        prs = github_client.search_issues(query)
+        if prs.totalCount == 0:
+            return {
+                "jira_summary": jira_info['summary'],
+                "jira_comments": jira_info['comments'],
+                "github_pr": "No corresponding GitHub PR found."
+            }
+
+        # 3. Get PR details from the first match
+        pr_number = prs[0].number
+        pr_details = get_github_pr_details(pr_number, repo_name)
+
+        # 4. Combine and return
+        return {
+            "jira_key": jira_key,
+            "jira_summary": jira_info['summary'],
+            "jira_comments": jira_info['comments'],
+            "github_pr_title": pr_details.get('title'),
+            "github_pr_description": pr_details.get('description'),
+            "github_pr_changed_files": pr_details.get('changed_files'),
+        }
+    except Exception as e:
+        return f"Error generating knowledge summary: {e}"
+
+@mcp.tool()
+def search_project_docs(query: str):
+    """
+    Search through static Mifos documentation and READMEs.
+    Use this for architectural questions or project setup guides.
+    """
+    # Simple implementation: search for keywords in .md files
+    docs_path = "./docs" # Or root if docs/ doesn't exist
+    results = []
+    if not os.path.exists(docs_path):
+        docs_path = "." # Fallback to root
+
+    for root, dirs, files in os.walk(docs_path):
+        for file in files:
+            if file.endswith(".md"):
+                with open(os.path.join(root, file), 'r', encoding='utf-8') as f:
+                    content = f.read()
+                    if query.lower() in content.lower():
+                        results.append(f"--- {file} ---\n{content[:500]}...")
+
+    return "\n".join(results) if results else "No matching documentation found."
+
+if __name__ == "__main__":
+    mcp.run()

Repo Clone Automation/repo_cloner.py

@@ -2,68 +2,75 @@
 import time
 import zipfile
 from selenium import webdriver
-from selenium.webdriver.chrome.service import Service
 from selenium.webdriver.common.by import By
 from selenium.webdriver.support.ui import WebDriverWait
 from selenium.webdriver.support import expected_conditions as EC
 
-# Set your download directory (absolute path)
-download_dir = os.path.abspath("downloads")
-if not os.path.exists(download_dir):
-    os.makedirs(download_dir)
+def clone_repository(repo_url, download_dir):
+    """
+    Clones a GitHub repository by downloading it as a ZIP file and extracting it.
+    """
+    # Set your download directory (absolute path)
+    if not os.path.exists(download_dir):
+        os.makedirs(download_dir)
 
-chrome_options = webdriver.ChromeOptions()
-prefs = {
-    "download.default_directory": download_dir,
-    "download.prompt_for_download": False,
-    "download.directory_upgrade": True,
-    "safebrowsing.enabled": True
-}
-chrome_options.add_experimental_option("prefs", prefs)
+    chrome_options = webdriver.ChromeOptions()
+    prefs = {
+        "download.default_directory": download_dir,
+        "download.prompt_for_download": False,
+        "download.directory_upgrade": True,
+        "safebrowsing.enabled": True
+    }
+    chrome_options.add_experimental_option("prefs", prefs)
 
-# Initialize the Chrome driver (ensure chromedriver is in your PATH)
-driver = webdriver.Chrome(options=chrome_options)
+    # Initialize the Chrome driver (ensure chromedriver is in your PATH)
+    driver = webdriver.Chrome(options=chrome_options)
 
-try:
-    # Navigate to the GitHub repository page
-    repo_url = "https://github.com/openMF/mifos-gazelle?tab=readme-ov-file"  # add ur repos, if needed i will write a function take multiple repo input to extract effectively
-    driver.get(repo_url)
-
-    # Wait until the "Code" button is clickable and click it.But here (//span[contains(text(), 'Code')]) isnt a unique locator so i have to use below provided one, this has tendency to change if github updates their frontend
-    wait = WebDriverWait(driver, 10)
-    code_button = wait.until(EC.element_to_be_clickable((By.XPATH, "//button[@data-variant='primary']//span[contains(@class, 'prc-Button-Label-pTQ')]")))
-    code_button.click()
-
-    # this is the download zip locator which is unique
-    download_zip = wait.until(EC.element_to_be_clickable((By.XPATH, "//span[contains(text(), 'Download ZIP')]")))
-    download_zip.click()
-
-    # Wait for the ZIP file to appear in the download folder (timeout after 60 seconds)
-    zip_filename = None
-    timeout = 60  # seconds
-    start_time = time.time()
-    while time.time() - start_time < timeout:
-        for filename in os.listdir(download_dir):
-            if filename.endswith(".zip"):
-                zip_filename = filename
+    try:
+        # Navigate to the GitHub repository page
+        driver.get(repo_url)
+
+        # Wait until the "Code" button is clickable and click it
+        wait = WebDriverWait(driver, 10)
+        code_button = wait.until(EC.element_to_be_clickable((By.XPATH, "//button[@data-variant='primary']//span[contains(@class, 'prc-Button-Label-pTQ')]")))
+        code_button.click()
+
+        # this is the download zip locator which is unique
+        download_zip = wait.until(EC.element_to_be_clickable((By.XPATH, "//span[contains(text(), 'Download ZIP')]")))
+        download_zip.click()
+
+        # Wait for the ZIP file to appear in the download folder
+        zip_filename = None
+        timeout = 60  # seconds
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            for filename in os.listdir(download_dir):
+                if filename.endswith(".zip"):
+                    zip_filename = filename
+                    break
+            if zip_filename:
                 break
-        if zip_filename:
-            break
-        time.sleep(1)
-
-    if not zip_filename:
-        print("Download timed out or ZIP file not found.")
-    else:
-        zip_path = os.path.join(download_dir, zip_filename)
-        print(f"Downloaded file: {zip_path}")
+            time.sleep(1)
 
-        # Unzip the downloaded archive into a subfolder called "extracted"
-        extract_dir = os.path.join(download_dir, "extracted")
-        if not os.path.exists(extract_dir):
-            os.makedirs(extract_dir)
-        with zipfile.ZipFile(zip_path, 'r') as zip_ref:
-            zip_ref.extractall(extract_dir)
-        print(f"Extracted ZIP contents to: {extract_dir}")
+        if not zip_filename:
+            print("Download timed out or ZIP file not found.")
+        else:
+            zip_path = os.path.join(download_dir, zip_filename)
+            print(f"Downloaded file: {zip_path}")
+
+            # Unzip the downloaded archive into a subfolder called "extracted"
+            extract_dir = os.path.join(download_dir, "extracted")
+            if not os.path.exists(extract_dir):
+                os.makedirs(extract_dir)
+            with zipfile.ZipFile(zip_path, 'r') as zip_ref:
+                zip_ref.extractall(extract_dir)
+            print(f"Extracted ZIP contents to: {extract_dir}")
+
+    finally:
+        driver.quit()
 
-finally:
-    driver.quit()
+if __name__ == "__main__":
+    # Original hardcoded values
+    repo_to_clone = "https://github.com/openMF/mifos-gazelle?tab=readme-ov-file"
+    download_destination = os.path.abspath("downloads")
+    clone_repository(repo_to_clone, download_destination)

Slack_scraper_bot/scripts/pii_remocval.py

@@ -1,7 +1,7 @@
-import scrubadub
-import scrubadub_spacy
 import re
 import sys
+import scrubadub
+import scrubadub_spacy
 
 def create_scrubber():
     scrubber = scrubadub.Scrubber()
@@ -19,7 +19,7 @@ def remove_user_tags(text):
     return re.sub(timestamp_user_pattern, lambda m: m.group(0).split('] User:')[0] + ']', text)
 
 def remove_name_lines(text):
-    name_pattern = r'^.*(?:my name is|I am|I\'m)\s+(?:{{NAME}}|[A-Z][a-z]+(?:\s+[A-Z][a-z]+)*).*$\n?'
+    name_pattern = r'^.*(?:my name is|I am|I\'m).*\n'
     return re.sub(name_pattern, '', text, flags=re.MULTILINE | re.IGNORECASE)
 
 def process_file(input_path, output_path):