# shengranhu/adas architecture

> Architecture diagram of the GitHub repository shengranhu/adas, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/shengranhu/adas
- Repository: https://github.com/shengranhu/adas

## Overview

ADAS is a research system for automatically inventing and evaluating agent designs. A user launches a domain-specific search; the search uses domain prompts to propose code, evaluates candidate agents on domain tasks, and records results for later iterations. The repository also includes cross-domain transfer-evaluation utilities. Search and evaluation are closely integrated in the sampled code, but unsampled domain implementations prevent confirming every internal call path; dashed links mark those uncertain relationships.

Main entry points: [`_arc/search.py`](https://github.com/shengranhu/adas/blob/main/_arc/search.py) (ARC Search), [`dataset`](https://github.com/shengranhu/adas/tree/main/dataset) (Benchmark Data), [`results`](https://github.com/shengranhu/adas/tree/main/results) (Search Archives), [`_drop/search.py`](https://github.com/shengranhu/adas/blob/main/_drop/search.py) (DROP Search)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_search["Agent Search"]
  node_arc_search["ARC Search<br/>[search.py]"]
  node_drop_search["DROP Search<br/>[search.py]"]
  node_gpqa_search["GPQA Search<br/>[search.py]"]
  node_mgsm_search["MGSM Search<br/>[search.py]"]
  node_mmlu_search["MMLU Search<br/>[search.py]"]
end

subgraph group_domains["Evaluation Domains"]
  node_arc_prompt["ARC Prompts<br/>[arc_prompt.py]"]
  node_drop_prompt["DROP Prompts<br/>[drop_prompt.py]"]
  node_gpqa_prompt["GPQA Prompts<br/>[gpqa_prompt.py]"]
  node_mgsm_prompt["MGSM Prompts<br/>[mgsm_prompt.py]"]
  node_mmlu_prompt["MMLU Prompts<br/>[mmlu_prompt.py]"]
  node_arc_utils["ARC Evaluation<br/>[utils.py]"]
  node_arc_select["ARC Selection<br/>[data_select.py]"]
end

subgraph group_transfer["Transfer Evaluation"]
  node_transfer_eval["Transfer Evaluators"]
end

subgraph group_state["Data and Results"]
  node_domain_data["Benchmark Data"]
  node_run_results["Search Archives"]
end

node_user(("Researcher"))

node_user -->|"launches"| node_arc_search
node_user -->|"launches"| node_drop_search
node_user -->|"launches"| node_gpqa_search
node_user -->|"launches"| node_mgsm_search
node_user -->|"launches"| node_mmlu_search
node_arc_search -.->|"uses prompts"| node_arc_prompt
node_drop_search -.->|"uses prompts"| node_drop_prompt
node_gpqa_search -.->|"uses prompts"| node_gpqa_prompt
node_mgsm_search -.->|"uses prompts"| node_mgsm_prompt
node_mmlu_search -.->|"uses prompts"| node_mmlu_prompt
node_arc_search -->|"evaluates with"| node_arc_utils
node_arc_search -.->|"reads tasks"| node_domain_data
node_drop_search -.->|"reads tasks"| node_domain_data
node_gpqa_search -.->|"reads tasks"| node_domain_data
node_mgsm_search -.->|"reads tasks"| node_domain_data
node_mmlu_search -.->|"reads tasks"| node_domain_data
node_arc_search -.->|"stores archive"| node_run_results
node_drop_search -.->|"stores archive"| node_run_results
node_gpqa_search -.->|"stores archive"| node_run_results
node_mgsm_search -.->|"stores archive"| node_run_results
node_mmlu_search -->|"stores archive"| node_run_results

click node_arc_search "https://github.com/shengranhu/adas/blob/main/_arc/search.py"
click node_drop_search "https://github.com/shengranhu/adas/blob/main/_drop/search.py"
click node_gpqa_search "https://github.com/shengranhu/adas/blob/main/_gpqa/search.py"
click node_mgsm_search "https://github.com/shengranhu/adas/blob/main/_mgsm/search.py"
click node_mmlu_search "https://github.com/shengranhu/adas/blob/main/_mmlu/search.py"
click node_arc_prompt "https://github.com/shengranhu/adas/blob/main/_arc/arc_prompt.py"
click node_drop_prompt "https://github.com/shengranhu/adas/blob/main/_drop/drop_prompt.py"
click node_gpqa_prompt "https://github.com/shengranhu/adas/blob/main/_gpqa/gpqa_prompt.py"
click node_mgsm_prompt "https://github.com/shengranhu/adas/blob/main/_mgsm/mgsm_prompt.py"
click node_mmlu_prompt "https://github.com/shengranhu/adas/blob/main/_mmlu/mmlu_prompt.py"
click node_arc_utils "https://github.com/shengranhu/adas/blob/main/_arc/utils.py"
click node_arc_select "https://github.com/shengranhu/adas/blob/main/_arc/data_select.py"
click node_domain_data "https://github.com/shengranhu/adas/tree/main/dataset"
click node_run_results "https://github.com/shengranhu/adas/tree/main/results"
click node_transfer_eval "https://github.com/shengranhu/adas/tree/main/_transfer_math"

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_arc_search,node_drop_search,node_gpqa_search,node_mgsm_search,node_mmlu_search toneBlue
class node_arc_prompt,node_drop_prompt,node_gpqa_prompt,node_mgsm_prompt,node_mmlu_prompt,node_arc_utils,node_arc_select toneAmber
class node_transfer_eval toneMint
class node_domain_data,node_run_results toneRose
class node_user toneIndigo
```

## Components

### Agent Search

- [ARC Search](https://github.com/shengranhu/adas/blob/main/_arc/search.py): `_arc/search.py`
- [DROP Search](https://github.com/shengranhu/adas/blob/main/_drop/search.py): `_drop/search.py`
- [GPQA Search](https://github.com/shengranhu/adas/blob/main/_gpqa/search.py): `_gpqa/search.py`
- [MGSM Search](https://github.com/shengranhu/adas/blob/main/_mgsm/search.py): `_mgsm/search.py`
- [MMLU Search](https://github.com/shengranhu/adas/blob/main/_mmlu/search.py): `_mmlu/search.py`

### Evaluation Domains

- [ARC Prompts](https://github.com/shengranhu/adas/blob/main/_arc/arc_prompt.py): `_arc/arc_prompt.py`
- [DROP Prompts](https://github.com/shengranhu/adas/blob/main/_drop/drop_prompt.py): `_drop/drop_prompt.py`
- [GPQA Prompts](https://github.com/shengranhu/adas/blob/main/_gpqa/gpqa_prompt.py): `_gpqa/gpqa_prompt.py`
- [MGSM Prompts](https://github.com/shengranhu/adas/blob/main/_mgsm/mgsm_prompt.py): `_mgsm/mgsm_prompt.py`
- [MMLU Prompts](https://github.com/shengranhu/adas/blob/main/_mmlu/mmlu_prompt.py): `_mmlu/mmlu_prompt.py`
- [ARC Evaluation](https://github.com/shengranhu/adas/blob/main/_arc/utils.py): `_arc/utils.py`
- [ARC Selection](https://github.com/shengranhu/adas/blob/main/_arc/data_select.py): `_arc/data_select.py`

### Transfer Evaluation

- [Transfer Evaluators](https://github.com/shengranhu/adas/tree/main/_transfer_math): `_transfer_math`

### Data and Results

- [Benchmark Data](https://github.com/shengranhu/adas/tree/main/dataset): `dataset`
- [Search Archives](https://github.com/shengranhu/adas/tree/main/results): `results`

### Other components

- **Researcher**

## Connections

- Researcher → ARC Search: launches (evidence: `README.md`)
- Researcher → DROP Search: launches (evidence: `README.md`)
- Researcher → GPQA Search: launches (evidence: `README.md`)
- Researcher → MGSM Search: launches (evidence: `README.md`)
- Researcher → MMLU Search: launches (evidence: `README.md`)
- ARC Search → ARC Prompts: uses prompts (evidence: `_arc/search.py`)
- DROP Search → DROP Prompts: uses prompts (evidence: `_drop/search.py`)
- GPQA Search → GPQA Prompts: uses prompts (evidence: `_gpqa/search.py`)
- MGSM Search → MGSM Prompts: uses prompts (evidence: `_mgsm/search.py`)
- MMLU Search → MMLU Prompts: uses prompts (evidence: `_mmlu/search.py`)
- ARC Search → ARC Evaluation: evaluates with (evidence: `_arc/search.py`)
- ARC Search → Benchmark Data: reads tasks
- DROP Search → Benchmark Data: reads tasks
- GPQA Search → Benchmark Data: reads tasks
- MGSM Search → Benchmark Data: reads tasks
- MMLU Search → Benchmark Data: reads tasks
- ARC Search → Search Archives: stores archive
- DROP Search → Search Archives: stores archive
- GPQA Search → Search Archives: stores archive
- MGSM Search → Search Archives: stores archive
- MMLU Search → Search Archives: stores archive (evidence: `_mmlu/search.py`)
