Proposal

The ability to invite others to my org/project is useful for collaboration.

Use-Case

No response

Is this a feature you are interested in implementing yourself?

Yes

Proposal

I want to be able to use Pezzo for monitoring and caching with LangChain.

Community - please thumbs up if you're interested in this feature.

We don't make use of the Next.js capabilities in the console app, therefore it does not make sense to use Next.

Additionally, we'd like to take advantage of runtime env variable injection (when deploying to different stages/environments) and this is not possible with Next.js as soon as SSR is enabled.

Make sure that for the Prompt model, name is a unique field (@unique). Display a friendly error message if it already exists.

In docker-compose.yaml, line 8, POSTGRES_USER should be postgres.

Proposal

Currently, all Console pages have the Pezzo title. This feature is about adding a title for all pages. The format should be as follows: {feature} - Pezzo.

Examples:

• Login page: Login - Pezzo
• Project Dashboard page: Dashboard - Pezzo
• Prompts page: Prompts - Pezzo
• Environments page: Environments - Pezzo
• ...etc

Use-Case

This will provide a better user experience, and make it much easier to understand which Pezzo tab is most relevant by providing context at a glance.

How to approach this

We have usehooks-ts nstalled. More specifically, you can use the useDocumntTitle hook to change the title per page.

All pages can be found in a component with a *Page.tsx file name. E.g. LoginPage.tsx, ProjectsPage.tsx and so on.

Proposal

Currently we do not provide the ability to remove an environment. In this feature, the following needs to be implemented:

Console

• Implement a delete button next to each environment (take an example from the organization members view)
• Upon clicking the button, a confirmation modal will appear with a message saying "Are you sure you want to delete the {{environment name}} environment? All associated data will be lost."
• Upon confirming, a GraphQL mutation deleteEnvironment will be sent

Server

• Change the Prisma schema to support cascade deletion of anything that has a relation to an environment. I believe this is currently only the PromptEnvironment model, which represents a deployment of a prompt version. Simply put, this means that whenever you delete an environment, you will also delete its deployment (PromptEnvironment instances). This can be automated in Prisma, see here.
• Implement the GraphQL mutation in the appropriate resolver/service (identity/environments.resolver.ts), validate that the user is an admin of their org (there are plenty of examples in the codebase).
• The environment will be deleted.

Use-Case

Ability to delete environments in case they were created accidentally or aren't needed anymore.

Note

Make sure you branch off of release/0.4.0, which is our next release.

We need a seed script that seeds some default data to a new installation.

• Environment: demo
• Prompt: GreetNewUser

Proposal

I would love to be able to add prompt test data and expected results for each prompt commit version. A lot of value in the ability to run that on a regular basis along with my other tests.

Use-Case

No response

Is this a feature you are interested in implementing yourself?

No

Proposal

It would be nice to add Azure OpenAI Completions to our project. It's widely used in business, so it could attract more users and keep us ahead as AI and ML become standard. Besides, implementation shouldn't differ that much and there are libs that already support OpenAI on azure platform.

Use-Case

Similar to OpenAI.

Is this a feature you are interested in implementing yourself?

Maybe

It would be good to have a demo app that uses AI (currently we're looking at OpenAI gpt-3.5-turbo chat completions API). This demo app can help onboard new Pezzo users.

Ideas are welcome! My initial idea was a simple Poem Writer app.

Report

In production, I am unable to save new API keys.

Expected behavior

The API key should be saved.

Steps to reproduce the problem

1. Go to an org page
2. Navigate to the "API Keys" tab
3. Modify the value of an API key and hit save

Logs (if applicable)

ncaught (in promise) TypeError: Cannot read properties of undefined (reading 'organizationId')
    at onClick (ProviderApiKeyListItem.tsx:61:31)
    at Y (button.js:134:55)
    at Object.$e (react-dom.production.min.js:54:317)
    at Be (react-dom.production.min.js:54:471)
    at react-dom.production.min.js:55:35
    at zn (react-dom.production.min.js:105:68)
    at Fn (react-dom.production.min.js:106:380)
    at react-dom.production.min.js:117:104
    at fu (react-dom.production.min.js:274:42)
    at Ie (react-dom.production.min.js:52:375)
onClick @ ProviderApiKeyListItem.tsx:61
Y @ button.js:134
$e @ react-dom.production.min.js:54
Be @ react-dom.production.min.js:54
(anonymous) @ react-dom.production.min.js:55
zn @ react-dom.production.min.js:105
Fn @ react-dom.production.min.js:106
(anonymous) @ react-dom.production.min.js:117
fu @ react-dom.production.min.js:274
Ie @ react-dom.production.min.js:52
Gn @ react-dom.production.min.js:109
Yt @ react-dom.production.min.js:74
qt @ react-dom.production.min.js:73

Pezzo version

0.3.0

How do you use Pezzo?

Pezzo Cloud

Proposal

simple query/function to get all prompts by the environment. if we have descriptions as well, on the client side it will be easy to know what the prompt will do when you execute it without even being in pezzo.

Use-Case

this would allow passing these as variables to pezzo functions and show all available prompts client side.

Is this a feature you are interested in implementing yourself?

yes

Report

Starting point
A basic prompt with a single variable

Inital test
A null value in user_message returns "Hello! How can I assist you today?"

Editing reqest

• Added "Alive? " to the reqeust
• Result gives a respose cosistent to what I would expect.

Using the variable to pass in "Alive?"

• "Alive?" does not seem to be passed to OpenAI

Expected behavior

I'm expecting the variable to be passed to the AI service. It does not apear that it is happening.

Steps to reproduce the problem

Logs (if applicable)

example

Pezzo version

No response

How do you use Pezzo?

None

The initial Alpha release of Pezzo contains over 10 database migrations. Most of these are experiments that have no value since no users have used Pezzo prior to the release.

Cleanup the migrations and start with a clean slate.

The github link on your website is broken. Points to https://github.com/pezzolabs.pezzo instead of https://github.com/pezzolabs/pezzo

Proposal

Add Support submenu with link to docs, Discord and useful links

Use-Case

No response

Is this a feature you are interested in implementing yourself?

Yes

Report

Running on MacOS I've been attempting to build the docker image provided in the install instructions and encountering an issue preventing it from starting correctly:

dependency failed to start: container pezzo-bold_shtern-pezzo-server-1 is unhealthy
exit status 1

Expected behavior

For docker desktop to compile and build

Steps to reproduce the problem

1. download repository via macOS terminal and git build
2. docker-compose build the application

Logs (if applicable)

ontainer pezzo-bold_shtern-supertokens-1  Created
 Container pezzo-bold_shtern-postgres-1  Created
 Container pezzo-bold_shtern-pezzo-server-1  Creating
 Container pezzo-bold_shtern-pezzo-server-1  Created
 Container pezzo-bold_shtern-pezzo-console-1  Creating
 Container pezzo-bold_shtern-pezzo-console-1  Created
 Container pezzo-bold_shtern-supertokens-1  Starting
 Container pezzo-bold_shtern-app-1  Starting
 Container pezzo-bold_shtern-postgres-1  Starting
 Container pezzo-bold_shtern-influxdb-1  Starting
 Container pezzo-bold_shtern-app-1  Started
 Container pezzo-bold_shtern-influxdb-1  Started
 Container pezzo-bold_shtern-supertokens-1  Started
 Container pezzo-bold_shtern-postgres-1  Started
 Container pezzo-bold_shtern-postgres-1  Waiting
 Container pezzo-bold_shtern-postgres-1  Healthy
 Container pezzo-bold_shtern-pezzo-server-1  Starting
 Container pezzo-bold_shtern-pezzo-server-1  Started
 Container pezzo-bold_shtern-pezzo-server-1  Waiting
 Container pezzo-bold_shtern-pezzo-server-1  Error
dependency failed to start: container pezzo-bold_shtern-pezzo-server-1 is unhealthy
exit status 1

Pezzo version

No response

How do you use Pezzo?

Docker Compose

Report

Updated to from previous releases

"@pezzo/client": "^0.2.0",
"@pezzo/integrations": "^0.1.8"

and when run build, getting an error

node_modules/@pezzo/client/src/client/Pezzo.d.ts:2:47 - error TS2307: Cannot find module '@pezzo/common' or its corresponding type declarations.

2 import type { CreatePromptExecutionDto } from "@pezzo/common";

Expected behavior

Expect not to get such a build issue

Steps to reproduce the problem

1. pull the latest version of client / integration packages and try to use it :)

Logs (if applicable)

node_modules/@pezzo/client/src/client/Pezzo.d.ts:2:47 - error TS2307: Cannot find module '@pezzo/common' or its corresponding type declarations.

2 import type { CreatePromptExecutionDto } from "@pezzo/common";

Pezzo version

client version 0.2.0

How do you use Pezzo?

Kubernetes Deployment

Pezzo console

Actual usage

Proposal

Currently, it is needed to manually run DB migrations using the Prisma CLI. However, when deploying Pezzo via Docker Compose, it is desired that the migrations will apply automatically when spinning up the services, so no manual intervention is needed.

Use-Case

No response

Is this a feature you are interested in implementing yourself?

Yes

Report

Expected behavior

I expect the variables to display the variable names upon loading the editor. Instead, I have to type something for the state to sync.

Steps to reproduce the problem

1. Create a prompt with variables
2. Commit a new version
3. Refresh
4. You will now see that the variables aren't shown

Logs (if applicable)

No response

Pezzo version

v0.1.4

How do you use Pezzo?

Pezzo Cloud

We need to have the ability to delete a prompt.

add a service method to prompts.service.ts to delete the prompt. You'l use Prisma to delete it (look at the other methods in the service).

Then add a deletePrompt mutation to the prompts.resolver.ts.

The Delete button should be displayed as follows:

There's already a prompt deletion confirmation modal implemened (check out apps/console/src/app/pages/prompts/[promptId]/index.tsx line 69). So all in all it's a fun task.

p.s. you'll need to adjust the Prisma Schema to cascade delete all dependants of a Prompt when you delete it. You can see how to do that here.

To support M1 users as well as non-ARM processors, the npx nx docker:build [project] target should support multiple platforms.

Report

Pezzo Client does not return latest deployed prompt. If there are multiple deployments to the same environment, the oldest is returned instead of the newest.

Expected behavior

Return the newest prompt version.

Steps to reproduce the problem

1. Create a promp
2. Deploy to an environment
3. Make a change, commit, deploy again
4. Consume the prompt with @pezzo/client

The 1st prompt is executed, not the latest one.

Logs (if applicable)

example

Pezzo version

v0.0.21

How do you use Pezzo?

None

When starting the app, the server throws an error due to generated GraphQL types not being available.
This is likely an issue with the order of execution of Nx commands.

Proposal

It would be valuable to have an auto-generated OpenAPI spec for Pezzo's REST API.

• REST is used exclusively by our SDK clients (currently JS and Python), and would be a great way for anyone to integrate, regardless of whether there is an SDK available or not.
• All REST endpoints are defined in NestJS controllers (*.controller.ts) in the apps/server directory.
• Solution must use @nestjs/swagger

Proposal

Depends on #190.

• Once we have an auto-generated OpenAPI spec (#190), we can insert it into our documentation site.
• Our docsite can be found in /docs.
• We use Mintlify. Mintlify has a feature for this. Read more about it here.

Use-Case

No response

Is this a feature you are interested in implementing yourself?

No

Proposal

• Compare results of a single prompt among many models
• Ability to add more models especially custom one. Something like a plain simple API request to a hosted endpoint.

Use-Case

New prompt engineers and companies have to iterate and switch regularly between models to test outputs of similar prompts. Instead a single interface can be developed in which new models can be added and their results could be compared in one go.

Is this a feature you are interested in implementing yourself?

No

Report

prisma -v
Environment variables loaded from .env
prisma : 5.3.1
@prisma/client : 5.1.1

Expected behavior

No response

Steps to reproduce the problem

Logs (if applicable)

example

Pezzo version

No response

How do you use Pezzo?

None

Proposal

Not entirely sure how difficult it would be to add prompt chains to this. But do feel system messages are a necessity, given the new versions of OpenAI API released this month. Would love to see these features implemented at some point.

Use-Case

Primary use case would be for prototyping more complicated AI products. Could allow for far more consistent responses. As far as I know there isn't anything like what I'm describing. Or at least I haven't come across any.

Is this a feature you are interested in implementing yourself?

No

Report

There is a broken link in README.md which makes it a bit hard to contribute :)

README excerpt:

Please check out CONTRIBUTING.md before contributing.

Expected behavior

I expect to see how one would contribute to this repo

Steps to reproduce the problem

1. open github.com
2. search for pezzo
3. try to contribute

Logs (if applicable)

example

Pezzo version

No response

How do you use Pezzo?

None

Proposal

The current client only supports typescript. I hope to provide clients in more languages so that users can start using pezzo.ai faster, especially shell and Python.

Use-Case

just like

# Fetch the prompt from Pezzo
prompt=$(curl "https://api.pezzo.ai/api/prompts/v2/deployment?name=<prompt_name>&environmentName=<env>&projectId=<project_id>" -H "Content-Type: application/json" -H "x-pezzo-api-key: <api_key>" | jq .content.prompt)

# Processing data
new_content=${prompt/{sentence}/hey}
json_data='{"model":"gpt-3.5-turbo","messages":[{"role":"user","content":""}],"temperature":1,"max_tokens":2560,"top_p":1,"frequency_penalty":0,"presence_penalty":0}'
data=$(echo "${json_data}" | jq --arg new_content "${new_content}" '.messages[0].content = $new_content')

# Use the OpenAI API as you normally would
curl https://api.openai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d "${data}"

Is this a feature you are interested in implementing yourself?

Maybe

Hello here,

I really appreciate your work on this amazing project. However, I am trying to run it locally and I am facing this error after Sign-in

WARNING in ../../node_modules/@remix-run/router/dist/router.js
Module Warning (from ../../node_modules/@nx/webpack/node_modules/source-map-loader/dist/cjs.js):
Failed to parse source map from 'router.js.map': SyntaxError: Unterminated string in JSON at position 37997

WARNING in ../../libs/common/src/index.ts 4:23-42
Should not import the named export 'version' (imported as 'versionJson') from default-exporting module (only default export is available soon)

WARNING in ./src/app/tailwind-overrides.css (./src/app/tailwind-overrides.css.webpack[javascript/auto]!=!../../node_modules/css-loader/dist/cjs.js??ruleSet[1].rules[6].oneOf[4].use[1]!../../node_modules/postcss-loader/dist/cjs.js??ruleSet[1].rules[6].oneOf[4].use[2]!./src/app/tailwind-overrides.css)
Module Warning (from ../../node_modules/postcss-loader/dist/cjs.js):
Warning

(2:3) Nested @tailwind rules were detected, but are not supported.
Consider using a prefix to scope Tailwind's classes: https://tailwindcss.com/docs/configuration#prefix
Alternatively, use the important selector strategy: https://tailwindcss.com/docs/configuration#selector-strategy

C:\Users\Salil_Apte\Documents\Development\pezzo>docker-compose up
[+] Running 0/2
 - pezzo-server Error                                                                                              3.3s
 - pezzo-console Error                                                                                             3.3s
Error response from daemon: pull access denied for pezzolabs/pezzo-console, repository does not exist or may require 'docker login': denied: requested access to the resource is denied

Not being able to proceed further. Guessing the console repo is private 👶

Proposal

A tab for prompts that shows the schema/types that the prompt should return, and is editable.

Use-Case

this makes it so you dont have to do any custom code or prompting this yourself, and makes it easy to add changes to what we are passing the prompt / getting back.

Bonus if we can make it typesafe in typescript such that if we are not passing the right things in pezzo client it shows a type error

Is this a feature you are interested in implementing yourself?

yes

The empty state is configured to show an empty state ("No executions") when the backend returns an empty array for data. This looks like this:

However, we also do not display the control that allow the user to adjust the time frame, granularity etc. in this state. Which means that if users adjust their time range to a range where there were 0 results, the empty state will show and the user cannot make any adjustments, and is therefore stuck in this state.

To fix this bug, simply make sure to always render the controls (in MetricProvider.tsx).

Report

The current configuration of the monorepo contains 2 circular dependencies.
This setup would cause unexpected results and potentially cause race build conditions

Expected behavior

Libraries must not depends on applications and applications should not depends on other applications.

The correct setup would be:
graphql <- console
graphql <- client <- console
graphql <- server

Steps to reproduce the problem

Check the nx graph running npx nx affected:graph and select all projects

Logs (if applicable)

No response

Pezzo version

No response

How do you use Pezzo?

None

The way Supertokens works, is that whenever you get an Unauthorized status code, it'll try to refresh your token automatically.

With GraphQL, every response is a 200 response, therefore this Supertokens mechanism does not work. We need to intercept GraphQL response errors, and if the code is UNAUTHORIZED, try to manually refresh the token. If that fails, we manually sign the user out (refresh token is likely expired).

Report

When creating a new variable in the prompt, all of the existing variables return to "null".

Additionally, after running the prompt test, any of the variables that are edited aren't saved and are also returned as "null".

Pezzo version

May 28th, 2023

How do you use Pezzo?

Docker Compose

Report

`vince.gauthier@Vince-Gauthier-MacBook-Pro:~/Desktop/pezzo$ npx dotenv-cli -e apps/server/.env -- npx prisma migrate deploy --schema apps/server/prisma/schema.prisma

Environment variables loaded from .env
Prisma schema loaded from apps/server/prisma/schema.prisma
Datasource "db": PostgreSQL database "pezzo", schema "public" at "localhost:5432"

11 migrations found in prisma/migrations

Error: P3009

migrate found failed migrations in the target database, new migrations will not be applied. Read more about how to resolve migration issues in a production database: https://pris.ly/d/migrate-resolve
The 20230519223302_add_projects migration started at 2023-06-15 18:12:37.505388 UTC failed with the following logs:

`

Expected behavior

the migration happens successfully

Steps to reproduce the problem

Logs (if applicable)

example

Pezzo version

No response

How do you use Pezzo?

Local Development Setup

Proposal

one docker-compose deploy all service open : 4200

Use-Case

No response

Is this a feature you are interested in implementing yourself?

No

Hey, After clocking the marked link:

I am getting page not found:

Report

我在创建 prompt 的时候 后台报了一个错误。日志如下
I reported an error in the background while creating prompt. Log as follows

Expected behavior

正常创建

Steps to reproduce the problem

1.创建一个项目
2.创建一个 prompt
3.前端显示 服务器内部错误。后端错误日志如下

1. Create a project
2. Create a prompt
3. The front-end displays an internal server error. Back-end error logs are as follows

Logs (if applicable)

pezzo-main-pezzo-server-1   | INFO [09:34:49.743] ([Redacted]): Creating prompt version
pezzo-main-pezzo-server-1   |     requestId: "bb395f32-cf12-4a19-adc0-75894de8189f"
pezzo-main-pezzo-server-1   |     userId: "b28762e0-6928-4862-8576-89c81d28c66c"
pezzo-main-pezzo-server-1   |     content: {
pezzo-main-pezzo-server-1   |       "prompt": "Generate {numFacts} facts about the following topic: \"{topic}\""
pezzo-main-pezzo-server-1   |     }
pezzo-main-pezzo-server-1   |     message: "01"
pezzo-main-pezzo-server-1   |     promptId: "clm4on97h0000qj2dfutag1br"
pezzo-main-pezzo-server-1   |     service: "OpenAIChatCompletion"
pezzo-main-pezzo-server-1   |     settings: {
pezzo-main-pezzo-server-1   |       "model": "gpt-3.5-turbo",
pezzo-main-pezzo-server-1   |       "temperature": 0.7,
pezzo-main-pezzo-server-1   |       "max_tokens": 256,
pezzo-main-pezzo-server-1   |       "top_p": 1,
pezzo-main-pezzo-server-1   |       "frequency_penalty": 0,
pezzo-main-pezzo-server-1   |       "presence_penalty": 0
pezzo-main-pezzo-server-1   |     }
pezzo-main-pezzo-server-1   | ERROR [09:34:49.760] ([Redacted]): Error creating prompt version
pezzo-main-pezzo-server-1   |     requestId: "bb395f32-cf12-4a19-adc0-75894de8189f"
pezzo-main-pezzo-server-1   |     userId: "b28762e0-6928-4862-8576-89c81d28c66c"
pezzo-main-pezzo-server-1   |     content: {
pezzo-main-pezzo-server-1   |       "prompt": "Generate {numFacts} facts about the following topic: \"{topic}\""
pezzo-main-pezzo-server-1   |     }
pezzo-main-pezzo-server-1   |     message: "Cannot read properties of undefined (reading 'track')"
pezzo-main-pezzo-server-1   |     promptId: "clm4on97h0000qj2dfutag1br"
pezzo-main-pezzo-server-1   |     service: "OpenAIChatCompletion"
pezzo-main-pezzo-server-1   |     settings: {
pezzo-main-pezzo-server-1   |       "model": "gpt-3.5-turbo",
pezzo-main-pezzo-server-1   |       "temperature": 0.7,
pezzo-main-pezzo-server-1   |       "max_tokens": 256,
pezzo-main-pezzo-server-1   |       "top_p": 1,
pezzo-main-pezzo-server-1   |       "frequency_penalty": 0,
pezzo-main-pezzo-server-1   |       "presence_penalty": 0
pezzo-main-pezzo-server-1   |     }
pezzo-main-pezzo-server-1   |     stack: "TypeError: Cannot read properties of undefined (reading 'track')\n    at AnalyticsService.trackEvent (/app/main.js:4807:28)\n    at PromptVersionsResolver.createPromptVersion (/app/main.js:7874:28)\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    at async target (/app/node_modules/@nestjs/core/helpers/external-context-creator.js:74:28)\n    at async /app/node_modules/@nestjs/core/helpers/external-proxy.js:9:24"
pezzo-main-pezzo-server-1   |     error: {}

Pezzo version

last

How do you use Pezzo?

Docker Compose

Report

I followed the getting started in the readme, at the last step I get this error:

 >  NX   Running target serve for project console and 1 task it depends on:

 —————————————————————————————————————————————————————————————————————————————————————————————————

> nx run common:prebuild

Syncing version
Synced version 0.3.3

> nx run console:serve:development

<i> [webpack-dev-server] Project is running at:
<i> [webpack-dev-server] Loopback: http://localhost:4200/, http://[::1]:4200/
<i> [webpack-dev-server] 404s will fallback to '/index.html'

>  NX  Web Development Server is listening at http://localhost:4200/


warn - The `content` option in your Tailwind CSS configuration is missing or empty.
warn - Configure your content sources or your generated CSS will be missing styles.
warn - https://tailwindcss.com/docs/content-configuration
<i> [webpack-dev-middleware] wait until bundle finished: /
Entrypoint main = runtime.js vendor.css vendor.js main.css main.js 7 auxiliary assets
chunk (runtime: runtime) main.css, main.js (main) 532 KiB (javascript) 1.42 KiB (css/mini-extract) [initial] [rendered]
chunk (runtime: runtime) runtime.js (runtime) 33.9 KiB [entry] [rendered]
chunk (runtime: runtime) vendor.css, vendor.js (vendor) (id hint: vendor) 15 MiB (javascript) 3.59 KiB (css/mini-extract) [initial] [rendered] split chunk (cache group: vendor) (name: vendor)

WARNING in ../../libs/common/src/index.ts 4:23-42
Should not import the named export 'version' (imported as 'versionJson') from default-exporting module (only default export is available soon)

ERROR in ./src/app/components/organizations/OrgRoleSelector.tsx 6:0-62
Module not found: Error: Can't resolve '../../../@generated/graphql/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/components/organizations'

ERROR in ./src/app/components/prompts/views/DashboardView.tsx 5:0-97
Module not found: Error: Can't resolve '../../../../@generated/graphql/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/components/prompts/views'

ERROR in ./src/app/components/prompts/views/PromptHistoryView.tsx 9:0-79
Module not found: Error: Can't resolve '../../../../@generated/graphql/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/components/prompts/views'

ERROR in ./src/app/graphql/definitions/mutations/api-keys.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/mutations'

ERROR in ./src/app/graphql/definitions/mutations/environments.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/mutations'

ERROR in ./src/app/graphql/definitions/mutations/organizations.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/mutations'

ERROR in ./src/app/graphql/definitions/mutations/prompt-environments.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/mutations'

ERROR in ./src/app/graphql/definitions/mutations/prompts.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/mutations'

ERROR in ./src/app/graphql/definitions/queries/api-keys.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/environments.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/metrics.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/organizations.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/projects.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/prompt-executions.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/prompts.tsx 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/graphql/definitions/queries/users.ts 3:0-57
Module not found: Error: Can't resolve '../../../../@generated/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/graphql/definitions/queries'

ERROR in ./src/app/lib/hooks/useCurrentOrgMembership.tsx 9:0-62
Module not found: Error: Can't resolve '../../../@generated/graphql/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/lib/hooks'

ERROR in ./src/app/lib/providers/MetricContext.tsx 8:0-66
Module not found: Error: Can't resolve '../../../@generated/graphql/graphql' in '/Users/mahmoudmabrouk/code/pezzo/apps/console/src/app/lib/providers'

webpack compiled with 18 errors and 1 warning (1dfd5a5ac8dd9d94)
ERROR in ./apps/console/src/app/components/api-keys/ProviderApiKeyListItem.tsx:8:43

localhost:4200 returns on the webbrowser
Cannot GET /

Expected behavior

to get the web app

Steps to reproduce the problem

Basically followed the getting started step by step

Logs (if applicable)

example

Pezzo version

latest

How do you use Pezzo?

Docker Compose

Report

Over time, services were added to the docker-compose.yaml file and it is now complex, and not as straightforward as it should be. Additionally, DB migrations do not automatically apply (#194).

Expected behavior

• Should be as straightforward as docker-compose up
• Should be documented well, both in README.md and ideally in the docsite as well.
• Should have the appropriate dependencies between services, init scripts and healthchecks.

Steps to reproduce the problem

No response

Logs (if applicable)

No response

Pezzo version

0.5.9

How do you use Pezzo?

Docker Compose

Report

I'm trying to enter my API key into the box:

I put that same API key into AI21 and it works.

Expected behavior

the api key for openai gets set.

Steps to reproduce the problem

1. set up project with docker on m1 mac pro
2. open console sign in
3. try to put API key for openai

Logs (if applicable)

example

Pezzo version

latest

How do you use Pezzo?

Docker Compose

Proposal

在配置文件中增加 OPENAI API BASH 属性，openai 对象 可以通过读取 这个属性配置 OPENAI API BASH

Add the 'OPENAI API BASH' property to the configuration file, and openai objects can configure 'OPENAI API BASH' by reading this property

Use-Case

在一些国家的网络环境中，openai 官网 是不允许访问的。增加这个功能可以使用某些代理网站或者搭建本地的 openai 环境完成 提示测试
In some countries, the openai website is not allowed to be accessed. This feature can be added by using some proxy website or setting up a local openai environment to complete prompt testing

如果有人遇到同样的问题（openai 网络环境不允许访问）。可以更改 server 容器中的 main.js:10935 行 加入以下代码
If someone encounters the same problem (the openai network environment does not allow access). You can change the main.js:10935 line in the server container to include the following code

configuration.baseURL = "xxxxxxxxxxxxx";
this.openai = new openai_1.default(configuration);
this.chat = new Chat(pezzo,this.openai)

Is this a feature you are interested in implementing yourself?

No

Proposal

In the ExecutorOptions needs to add the organization field and use it in the initialization of OpenAIApi Configuration https://github.com/pezzolabs/pezzo/blob/main/libs/integrations/src/lib/integrations/openai/executor.ts#L12 and https://github.com/pezzolabs/pezzo/blob/main/libs/integrations/src/lib/integrations/openai/sdk.ts#L4

Use-Case

We are using multiple organizations in OpenAI :)

Is this a feature you are interested in implementing yourself?

Maybe

does not affect functionality, but maybe something in cache causes this.