# divya-pal4/code-review-openenv architecture

> Architecture diagram of the GitHub repository divya-pal4/code-review-openenv, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/divya-pal4/code-review-openenv
- Repository: https://github.com/divya-pal4/code-review-openenv

## Overview

Code Review OpenEnv is a gym-style benchmark for agents to review code snippets and receive task-specific feedback and scores. The principal workflow is: an agent resets the environment, receives a task observation, submits a review comment, and gets the next observation and grade. The baseline inference client optionally generates comments with an external language model. The README describes test-based grading, but the sampled grader implements keyword matching only; the excerpts do not show hidden-test execution.

Main entry points: [`server/app.py`](https://github.com/divya-pal4/code-review-openenv/blob/main/server/app.py) (REST API), [`env/environment.py`](https://github.com/divya-pal4/code-review-openenv/blob/main/env/environment.py) (Task environment), [`inference.py`](https://github.com/divya-pal4/code-review-openenv/blob/main/inference.py) (Baseline client), [`env/grader.py`](https://github.com/divya-pal4/code-review-openenv/blob/main/env/grader.py) (Keyword grader)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_client["Agent clients"]
  node_baseline["Baseline client<br/>[inference.py]"]
end

subgraph group_api["Environment API"]
  node_shim["App shim<br/>[app.py]"]
  node_server["REST API<br/>[app.py]"]
end

subgraph group_domain["Review environment"]
  node_environment["Task environment<br/>[environment.py]"]
  node_taskbank["Review tasks<br/>[environment.py]"]
  node_models["Action and observation models<br/>[models.py]"]
end

subgraph group_grading["Scoring"]
  node_grader["Keyword grader<br/>[grader.py]"]
  node_taskgraders["Task grader adapters"]
end

node_agent(("Review agent"))
node_llm(("Language model"))

node_agent -->|"uses REST"| node_server
node_baseline -->|"calls reset and step"| node_server
node_baseline -->|"requests comments"| node_llm
node_shim -->|"exports app"| node_server
node_server -->|"resets and steps"| node_environment
node_server -->|"reads state"| node_environment
node_server -->|"grades action"| node_grader
node_environment -->|"constructs observations"| node_models
node_environment -->|"selects task"| node_taskbank
node_taskgraders -->|"delegates grading"| node_grader

click node_baseline "https://github.com/divya-pal4/code-review-openenv/blob/main/inference.py"
click node_shim "https://github.com/divya-pal4/code-review-openenv/blob/main/app.py"
click node_server "https://github.com/divya-pal4/code-review-openenv/blob/main/server/app.py"
click node_environment "https://github.com/divya-pal4/code-review-openenv/blob/main/env/environment.py"
click node_taskbank "https://github.com/divya-pal4/code-review-openenv/blob/main/env/environment.py"
click node_models "https://github.com/divya-pal4/code-review-openenv/blob/main/env/models.py"
click node_grader "https://github.com/divya-pal4/code-review-openenv/blob/main/env/grader.py"
click node_taskgraders "https://github.com/divya-pal4/code-review-openenv/tree/main/graders"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_baseline,node_agent toneBlue
class node_shim,node_server toneAmber
class node_environment,node_taskbank,node_models toneMint
class node_grader,node_taskgraders toneRose
class node_llm toneIndigo
```

## Components

### Agent clients

- [Baseline client](https://github.com/divya-pal4/code-review-openenv/blob/main/inference.py): `inference.py`

### Environment API

- [App shim](https://github.com/divya-pal4/code-review-openenv/blob/main/app.py): `app.py`
- [REST API](https://github.com/divya-pal4/code-review-openenv/blob/main/server/app.py): `server/app.py`

### Review environment

- [Task environment](https://github.com/divya-pal4/code-review-openenv/blob/main/env/environment.py): `env/environment.py`
- [Review tasks](https://github.com/divya-pal4/code-review-openenv/blob/main/env/environment.py): `env/environment.py`
- [Action and observation models](https://github.com/divya-pal4/code-review-openenv/blob/main/env/models.py): `env/models.py`

### Scoring

- [Keyword grader](https://github.com/divya-pal4/code-review-openenv/blob/main/env/grader.py): `env/grader.py`
- [Task grader adapters](https://github.com/divya-pal4/code-review-openenv/tree/main/graders): `graders`

### Other components

- **Review agent**
- **Language model**

## Connections

- Review agent → REST API: uses REST (evidence: `README.md`)
- Baseline client → REST API: calls reset and step (evidence: `inference.py`)
- Baseline client → Language model: requests comments (evidence: `inference.py`)
- App shim → REST API: exports app (evidence: `app.py`)
- REST API → Task environment: resets and steps (evidence: `server/app.py`)
- REST API → Task environment: reads state (evidence: `server/app.py`)
- REST API → Keyword grader: grades action (evidence: `server/app.py`)
- Task environment → Action and observation models: constructs observations (evidence: `env/environment.py`)
- Task environment → Review tasks: selects task (evidence: `env/environment.py`)
- Task grader adapters → Keyword grader: delegates grading (evidence: `graders/task_1.py`)
