DevIOの新着記事がメディアポリシーに準拠しているかBedrockで自動レビューしてみた

import boto3
from datetime import datetime, timedelta, timezone
import feedparser
import os

sqs = boto3.client('sqs')
JST = timezone(timedelta(hours=+9))
feed_url = 'https://dev.classmethod.jp/feed/'
queue_url = os.environ['QUEUE_URL']

def get_feed_entries():
    updated_since = datetime.now(JST) - timedelta(hours=1)
    feed = feedparser.parse(feed_url)
    new_entries = [
        entry for entry in feed.entries
        if datetime(*entry.updated_parsed[:6], tzinfo=timezone.utc)
        .astimezone(JST) > updated_since
    ]
    return new_entries

def lambda_handler(event, context):

    new_entries = get_feed_entries()
    entries = [
        {
            "Id": entry["id"],
            "MessageBody": entry['link']
        } for entry in  new_entries
    ]
    if len(entries) == 0:
        print('新着記事が見つからなかったため処理を終了します')
        return
    sqs.send_message_batch(
        QueueUrl=queue_url,
        Entries=entries
    )

from aws_lambda_powertools import Logger
import boto3
import os
from slack_sdk import WebClient

from functions import review_and_notify_slack

logger = Logger()
ssm = boto3.client("ssm")

ssm_res = ssm.get_parameter(Name="/blog-review/prd/slack-bot-token", WithDecryption=True)
slack_token = ssm_res['Parameter']['Value']
slack_channel = os.environ["SLACK_CHANNEL_ID"]
slack_client = WebClient(token=slack_token)
bedrock_runtime = boto3.client("bedrock-runtime")
system_prompt = '''
あなたは企業ブログのレビュワーです

ブログ内に不適切な表現がないかチェックする必要があります。

...略
'''

def lambda_handler(event, context):

    records = event['Records']
    batch_item_failures = []
    response = {}

    for record in records:

        try:
            review_and_notify_slack(
                url=record['body'],
                system_prompt=system_prompt,
                bedrock_runtime=bedrock_runtime,
                slack_client=slack_client,
                slack_channel=slack_channel,
            )
        except Exception as e:
            logger.error({
                "message": "Failed to process record",
                "record": record,
                "error": str(e)
            })
            batch_item_failures.append({"itemIdentifier": record['messageId']})

    response["batchItemFailures"] = batch_item_failures
    return response

import json
import requests
from bs4 import BeautifulSoup
from markdownify import markdownify
from mypy_boto3_bedrock_runtime.client import BedrockRuntimeClient
from slack_sdk import WebClient

def review_and_notify_slack(
    url: str,
    system_prompt: str,
    bedrock_runtime: BedrockRuntimeClient,
    slack_client: WebClient,
    slack_channel: str,
):
    res = requests.get(url)
    soup = BeautifulSoup(res.text, "html.parser")
    article = soup.find("article")
    md_article = markdownify(article.prettify())

    model_id = "anthropic.claude-3-haiku-20240307-v1:0"
    review_content = f"""
以下ブログのレビューお願いします

```
{md_article}
```
    """

    res = bedrock_runtime.invoke_model(
        modelId=model_id,
        body=json.dumps(
            {
                "anthropic_version": "bedrock-2023-05-31",
                "max_tokens": 2000,
                "system": system_prompt,
                "messages": [
                    {
                        "role": "user",
                        "content": [{"type": "text", "text": review_content}],
                    }
                ],
            }
        ),
    )

    response_body = json.loads(res.get("body").read())
    print(response_body)

    slack_client.chat_postMessage(
        channel=slack_channel,
        blocks=[
            {
                "type": "section",
                "text": {"type": "mrkdwn", "text": "以下のブログをレビューしました"},
            },
            {"type": "section", "text": {"type": "mrkdwn", "text": url}},
            {"type": "divider"},
            {
                "type": "section",
                "text": {"type": "mrkdwn", "text": response_body["content"][0]["text"]},
            },
            {"type": "divider"},
        ],
    )

AWSTemplateFormatVersion: '2010-09-09'
Transform: AWS::Serverless-2016-10-31
Parameters:
  SlackChannelId:
    Type: String
    Description: Slack Channel ID
Globals:
  Function:
    Runtime: python3.12
    Timeout: 300
    MemorySize: 128
    Tracing: Active
  Api:
    TracingEnabled: true
Resources:
  BlogUrlQueue:
    Type: AWS::SQS::Queue
    Properties:
      VisibilityTimeout: 330
  BlogUrlQueuePolicy: 
    Type: AWS::SQS::QueuePolicy
    Properties: 
      PolicyDocument: 
        Statement: 
          - Sid: allo-sqs
            Effect: Allow
            Principal: "*"
            Action: "SQS:*"
            Resource: !GetAtt BlogUrlQueue.Arn
      Queues:
        - !Ref BlogUrlQueue
  BlogFeedFunc:
    Type: AWS::Serverless::Function
    Properties:
      CodeUri: blog-feed/
      Handler: app.lambda_handler
      Architectures:
      - x86_64
      Timeout: 10
      Environment:
        Variables:
          QUEUE_URL: !GetAtt BlogUrlQueue.QueueUrl
      Events:
        BlogFeedSchedule:
          Type: Schedule
          Properties:
            Schedule: cron(0 * * * ? *)
            Enabled: true
  BlogReviewFunc:
    Type: AWS::Serverless::Function
    Properties:
      CodeUri: blog-review/
      Handler: app.lambda_handler
      Architectures:
      - x86_64
      Role: !GetAtt BlogReviewFuncRole.Arn
      Events:
        SQSEvent:
          Type: SQS
          Properties:
            Queue: !GetAtt BlogUrlQueue.Arn
            BatchSize: 10
            FunctionResponseTypes:
              - ReportBatchItemFailures            
      Environment:
        Variables:
          SLACK_CHANNEL_ID: !Ref SlackChannelId
  BlogReviewFuncRole:
    Type: "AWS::IAM::Role"
    Properties:
      RoleName: bedrock-slack-backlog-rag-app-lambda-role
      AssumeRolePolicyDocument:
        Version: "2012-10-17"
        Statement:
          - Effect: Allow
            Principal:
              Service: lambda.amazonaws.com
            Action: sts:AssumeRole
      Policies:
        - PolicyName: allow-bedrock
          PolicyDocument:
            Version: "2012-10-17"
            Statement:
              - Effect: Allow
                Action:
                  - bedrock:InvokeAgent
                  - bedrock:InvokeModel
                  - bedrock:Retrieve
                  - bedrock:InvokeModelWithResponseStream
                Resource: "*"
        - PolicyName: allow-ssm
          PolicyDocument:
            Version: "2012-10-17"
            Statement:
              - Effect: Allow
                Action:
                  - ssm:GetParameter
                Resource: "*"                
        - PolicyName: allow-logs
          PolicyDocument:
            Version: '2012-10-17'
            Statement:
            - Effect: Allow
              Action:
              - logs:CreateLogGroup
              - logs:CreateLogStream
              - logs:PutLogEvents
              Resource: '*'

aws ssm  put-parameter --name '/blog-review/prd/slack-bot-token' --value 'Slackのボット用トークン' --type SecureString