Merge pull request #62 from Significant-Gravitas/ntindle/agpt-400-upd…

…ate-post-spec-to-schedule-a-background-task-for

[Needed for Discord bot] Segment Interview and Spec Generation

README.md

@@ -6,7 +6,6 @@
 
 The Codex System is an innovative coding agent designed to streamline the software development process. It consists of four key sub-agents, each specialized in a different aspect of software development. These sub-agents work in harmony to ensure efficient and effective delivery of software applications. This README provides an overview of the Codex System and its components.
 
-
 ## Getting Started
 
 Welcome to the initial setup guide for your project. Follow these easy steps to get everything up and running.
@@ -167,26 +166,26 @@ sequenceDiagram
       User->>API: Modifies App Spec Element
       API->>User: Returns a new App Spec ID
     end
-    
+
     User->>API: Request App Spec ID is Developed
     API->>+Codex: Runs Architect and Develop Flow
 
     Codex->>-User: Returns Developed App ID
-    
+
     User->>API: Requests Deveoped App ID is deployed
     Note right of API: When we have added deployment flow
 
     API->>+Codex: Runs Deployment Flow
     Codex-->>-User: Returns Deployment details (id, url etc)
 
-    
+
     User->>API: Requests Code for Developed App ID
     Note right of API: During Beta
 
     API->>+Codex: Runs Package flow
 
     Codex-->>-User: Returns Zipfile
-    
+
 ```
 
 ## Workflow
@@ -199,7 +198,6 @@ sequenceDiagram
 
 4. **Deployment**: Once the coding is complete, the Deploy sub-agent takes over to package, compile, and deploy the application to the desired environment.
 
-
 ```mermaid
 sequenceDiagram
     participant User
@@ -221,14 +219,12 @@ sequenceDiagram
     Deploy->>+User: Deploy to Production
 ```
 
-
-##  Database Schema
+## Database Schema
 
 <p align="center">
     <img src="docs/mermaid.png" alt="Mermaid Diagram of Schema" style="border-radius: 50px;" width="600">
 </p>
 
-
 The schema revolves around key models:
 
 - CodeGraph: Represents the logic and structure of code as graphs, linked to function definitions and database schemas.
@@ -333,9 +329,9 @@ erDiagram
     }
 ```
 
-## Useful commands 
-> docker buildx build --platform linux/amd64 -t gcr.io/agpt-dev/mvp/codegen . --push
+## Useful commands
 
+> docker buildx build --platform linux/amd64 -t gcr.io/agpt-dev/mvp/codegen . --push
 
 ## Prisma with Python: Quick Setup and Usage Guide
 
@@ -344,18 +340,21 @@ Prisma is an open-source database toolkit that simplifies database access and ma
 ### 1. Setting Up Prisma
 
 #### Prerequisites:
+
 - Node.js installed (for Prisma CLI)
 - Python environment setup
 
 #### Steps:
 
 1. **Install Prisma CLI**:
+
    - Use npm to install Prisma globally:
      ```bash
      npm install -g prisma
      ```
 
 2. **Initialize Prisma in Your Project**:
+
    - Navigate to your Python project directory and initialize Prisma:
      ```bash
      prisma init
@@ -410,6 +409,7 @@ Prisma is an open-source database toolkit that simplifies database access and ma
 ### 5. Using Prisma with Python
 
 - Since Prisma Client is native to JavaScript/TypeScript, using it in Python requires a workaround. You can execute Prisma Client through a child process. For example:
+
   ```python
   import subprocess
   import json
@@ -435,4 +435,4 @@ Prisma is an open-source database toolkit that simplifies database access and ma
 
 ### Conclusion
 
-This cheat sheet covers the basics of setting up Prisma in a Python project and performing essential database operations. Remember that using Prisma with Python is less straightforward than with JavaScript/TypeScript, and it may require additional setup and handling. For more detailed information, refer to the [Prisma Documentation](https://www.prisma.io/docs/).
+This cheat sheet covers the basics of setting up Prisma in a Python project and performing essential database operations. Remember that using Prisma with Python is less straightforward than with JavaScript/TypeScript, and it may require additional setup and handling. For more detailed information, refer to the [Prisma Documentation](https://www.prisma.io/docs/).

codex/api_model.py

@@ -3,15 +3,19 @@
 from typing import List, Optional
 
 from prisma.enums import Role
-from prisma.models import Specification
+from prisma.models import Question, Specification
 from pydantic import BaseModel, Field
 
+from codex.interview.model import InterviewMessage, InterviewMessageWithResponse
+
 logger = logging.getLogger(__name__)
 
 
 class Identifiers(BaseModel):
     user_id: str
+    cloud_services_id: str
     app_id: str
+    interview_id: str | None = None
     spec_id: str | None = None
     completed_app_id: str | None = None
     deployment_id: str | None = None
@@ -80,6 +84,7 @@ class ApplicationResponse(BaseModel):
     updatedAt: datetime
     name: str
     userid: str
+    cloud_services_id: str
 
 
 class ApplicationsListResponse(BaseModel):
@@ -124,6 +129,25 @@ class APIRouteSpecModel(BaseModel):
     responseObject: Optional[ResponseObjectModel] = None
 
 
+class InterviewCreate(BaseModel):
+    name: str
+    task: str
+
+
+class InterviewResponse(BaseModel):
+    id: str = Field(..., description="The unique identifier of the interview")
+    finished: bool = Field(bool, description="Whether the interview is finished")
+    uses: List[Question] = Field(
+        ..., description="The list of messages in the interview"
+    )
+
+
+class InterviewNextCreate(BaseModel):
+    uses: List[InterviewMessageWithResponse] = Field(
+        ..., description="The list of messages in the interview"
+    )
+
+
 class SpecificationCreate(BaseModel):
     description: str
 

codex/app.py

@@ -6,6 +6,7 @@
 from codex.api import core_routes
 from codex.deploy.routes import deployment_router
 from codex.develop.routes import delivery_router
+from codex.interview.routes import interview_router
 from codex.requirements.routes import spec_router
 
 logger = logging.getLogger(__name__)
@@ -23,6 +24,7 @@
 
 
 app.include_router(core_routes, prefix="/api/v1")
+app.include_router(interview_router, prefix="/api/v1")
 app.include_router(spec_router, prefix="/api/v1")
 app.include_router(delivery_router, prefix="/api/v1")
 app.include_router(deployment_router, prefix="/api/v1")

codex/common/ai_block.py

@@ -2,7 +2,7 @@
 import logging
 import os
 import pathlib
-from typing import Any
+from typing import Any, Callable, Optional, Type
 
 from jinja2 import Environment, FileSystemLoader
 from openai import AsyncOpenAI
@@ -11,7 +11,7 @@
 from prisma.enums import DevelopmentPhase
 from prisma.fields import Json
 from prisma.models import LLMCallAttempt, LLMCallTemplate
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 
 from codex.api_model import Identifiers
 from codex.common.ai_model import OpenAIChatClient
@@ -393,3 +393,11 @@ async def list_items(self, query_params: dict):
             query_params (dict): _description_
         """
         raise NotImplementedError("List Items Method not implemented")
+
+
+class Tool(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+    name: str
+    description: str
+    func: Optional[Callable[..., str]] = None
+    block: Type[AIBlock]

codex/common/test_const.py

@@ -2,6 +2,15 @@
 
 user_id_1 = "123e4567-e89b-12d3-a456-426614174000"
 user_id_2 = "456e7890-f12b-23d4-b567-537625285000"
+user_id_3 = "789a1234-c56d-67e8-c901-637636396000"
+
+discord_id_1 = "698489167281127474"
+discord_id_2 = "1188760123011776594"
+discord_id_3 = "1113853244066779156"
+
+cloud_services_id_1 = "9c7c6f29-c15e-40c9-afb3-c2c94f621a5c"
+cloud_services_id_2 = "78f3b489-e99e-4e50-a3aa-e78f8ad9b5b2"
+cloud_services_id_3 = "44286240-d819-4c57-beef-0bc815ca1934"
 
 app_id_1 = "789e0123-g34c-45d6-e789-012345678900"
 app_id_2 = "901f2345-h56i-67j8-k901-l23456789000"
@@ -17,6 +26,24 @@
 app_id_12 = "901p2345-r67s-78t9-u012-v34567890100"
 app_id_13 = "012q3456-s78t-89u0-v123-w45678901200"
 
+interview_id_1 = "d5cf71f4-068c-44aa-97c1-db864526d0cf"
+interview_id_2 = "24fefb91-5eae-472e-ab15-0c1ad158bea3"
+interview_id_3 = "51de45a0-d00d-40c6-9494-fc0941288a55"
+interview_id_4 = "571c30fc-5b22-4da5-8558-9b98e6da4dd2"
+interview_id_5 = "6aecf5eb-ea4c-4ba7-96d7-8847a41af480"
+interview_id_6 = "d1f0fba7-15e9-41a3-83e2-093d1f64138a"
+interview_id_7 = "e5c3386d-814c-4108-b09c-10e807238c8a"
+interview_id_8 = "11d9607e-3638-48c5-baac-6f0f9d96fb42"
+interview_id_9 = "2b5ccea7-347b-4940-bfe6-ae999fe874fc"
+interview_id_10 = "400a8c69-f448-4481-a996-134abb9fa1c0"
+interview_id_11 = "a0248eaf-3cac-4c9a-8757-fdda9e840a7c"
+interview_id_12 = "daf074f4-eac9-40c6-a6dd-f928aa6717ab"
+interview_id_13 = "b2ca05ca-3a41-4a89-9afd-7bd48eb170bd"
+
 
-identifier_1 = Identifiers(user_id=user_id_1, app_id=app_id_1)
-identifier_2 = Identifiers(user_id=user_id_2, app_id=app_id_2)
+identifier_1 = Identifiers(
+    user_id=user_id_1, app_id=app_id_1, cloud_services_id=cloud_services_id_1
+)
+identifier_2 = Identifiers(
+    user_id=user_id_2, app_id=app_id_2, cloud_services_id=cloud_services_id_2
+)