M-M-Hackathon

Running the api app

Navigate to the api directory
run npm i and then npm start

OpenAI API endpoints

inputPrompt (post)

• "inputPrompt" enpoint is used for general GPT inputs
• hosted at localhost:9000/openAI/inputPrompt
• input: { prompt : "The general instruction for the bot to follow", input:"The input file or text for the bot to respond to"}
• output: { role: "role of bot", content : "main response from the api"}

---

Client Help (post)

• "clientHelp" endpoint takes a file and intrepets it for the client
• hosted at localhost:9000/openAI/clientHelp
• input : {input:"input file"}
• output: { role: "role of bot", content : "main response from the api"}

---

lawyer Help (post)

• "lawyerHelp" endpoint takes a file and intrepets it for a lawyer
• hosted at localhost:9000/openAI/lawyerHelp
• input : {input:"input file"}
• output: { role: "role of bot", content : "main response from the api"}

---

Monkey Mode (post)

• "monkeyMode"
• hosted at localhost:9000/openAI/monkeyMode

---

Vector DB endpoints

add doc

• used to add documents to the database
• hosted at localhost:8080/vector/create
• input{ "class":"Document", "properties":{ "content": "hello" } }
• output {"class": "Document", "creationTimeUnix": 1682792671119, "id": "0d5425f2-139f-4fdf-829b-3516a0a784db", "lastUpdateTimeUnix": 1682792671119, "properties": { "content": "hello" }, "vector": [ -0.01775357, 0.00250483 ] }

Vector DB endpoints

add doc (post)

• used to add documents to the database
• hosted at localhost:8080/vector/create
• input{ "class":"Document", "properties":{ "content": "hello" } }
• output {"class": "Document", "creationTimeUnix": 1682792671119, "id": "0d5425f2-139f-4fdf-829b-3516a0a784db", "lastUpdateTimeUnix": 1682792671119, "properties": { "content": "hello" }, "vector": [ -0.01775357, 0.00250483 ] }

---

get all (get)

• gets all documents in the DB
• hosted at localhost:8080/vector/getAll
• no input
• outputs an array of docs