Clarify task examples and auxiliary tables

README.md

@@ -32,26 +32,51 @@ configs:
 
 # AsynCodeBench
 
-AsynCodeBench evaluates coding agents under five execution protocols while holding the task set and evaluation contracts fixed. This dataset card packages the metadata for the stable **v0.4.2** release: **19 tasks**, **95 task-protocol scenarios**, and **52 annotated dependency points**.
+AsynCodeBench evaluates coding agents under five execution protocols while holding the task set and evaluation contracts fixed. The stable **v0.4.2** release contains **19 coding tasks**. Start with the `tasks` configuration; each row is one real benchmark task.
 
-The executable harness, manifests, container references, evaluation commands, and Quick Start are versioned in the [official GitHub repository](https://github.com/KaituoZhang/AsynCodeBench/tree/2a639b049f249e79eb2031876bc06a110be9ac3a). The Hugging Face dataset is a query-friendly index of that release; it does not redistribute upstream source repositories or container images.
+[**Browse the 19 task examples in Dataset Viewer →**](https://huggingface.co/datasets/PhilipZhang/AsynCodeBench/viewer/tasks/test)
 
-## Configurations
+The executable harness, manifests, container references, evaluation commands, and Quick Start are versioned in the [official GitHub repository](https://github.com/KaituoZhang/AsynCodeBench/tree/2a639b049f249e79eb2031876bc06a110be9ac3a). Source repositories are materialized at each row's `base_revision` inside the recorded container image; the Parquet files therefore do not duplicate complete source checkouts or container layers.
 
-| Configuration | Rows | Description |
-|---|---:|---|
-| `tasks` | 19 | Task statements, source revisions, tests, container digests, and manifest provenance |
-| `scenarios` | 95 | Five official protocol conditions for every task |
-| `dependencies` | 52 | Producer-consumer dependency contracts and probe tests |
-
-All configurations use the `test` split because this is an evaluation benchmark.
+## Start here: one task example
 
-## Load the data
+The first `tasks` row is `asyncodebench:cachetools`. It contains the full coding prompt in `problem_statement`, the immutable source revision, evaluator command, test targets, and pinned container image/digest.
 
 ```python
 from datasets import load_dataset
 
-tasks = load_dataset("PhilipZhang/AsynCodeBench", "tasks", split="test", revision="v0.4.2")
+tasks = load_dataset(
+    "PhilipZhang/AsynCodeBench",
+    "tasks",
+    split="test",
+    revision="v0.4.2",
+)
+example = tasks[0]
+print(example["task_id"])             # asyncodebench:cachetools
+print(example["problem_statement"])   # complete task prompt
+print(example["container_image"])     # docker.io/wentingzhao/cachetools:v0
+print(example["dependency_count"])    # 5
+```
+
+## What the three configurations mean
+
+| Configuration | Rows | Purpose |
+|---|---:|---|
+| **`tasks`** | **19** | **Primary dataset. One row per coding task: prompt, source revision, tests, evaluator, and container. Most users only need this.** |
+| `scenarios` | 95 | Auxiliary protocol table: 19 tasks × 5 execution protocols. These are run conditions, not 95 additional coding tasks. |
+| `dependencies` | 52 | Auxiliary annotation table: producer-consumer contracts inside the 19 tasks, with probe tests for ADPR, DRS, CAIL, and SAD metrics. These are not additional tasks. |
+
+All configurations use the `test` split because every row belongs to the evaluation benchmark.
+
+- [Browse task examples](https://huggingface.co/datasets/PhilipZhang/AsynCodeBench/viewer/tasks/test)
+- [Browse protocol conditions](https://huggingface.co/datasets/PhilipZhang/AsynCodeBench/viewer/scenarios/test)
+- [Browse dependency annotations](https://huggingface.co/datasets/PhilipZhang/AsynCodeBench/viewer/dependencies/test)
+
+## Load auxiliary analysis tables
+
+You do not need these tables to enumerate the 19 coding tasks. Load them when comparing protocols or computing dependency-aware metrics:
+
+```python
 scenarios = load_dataset("PhilipZhang/AsynCodeBench", "scenarios", split="test", revision="v0.4.2")
 dependencies = load_dataset("PhilipZhang/AsynCodeBench", "dependencies", split="test", revision="v0.4.2")
 ```

checksums.json

@@ -1,6 +1,6 @@
 {
   "files": {
-    "README.md": "5abbd95b3d00ce4adac306481915e287ed1f338057340745b24d210a5f689464",
+    "README.md": "1cdba53b668ac408040f2e51878d97419030879b5900847d823b0ba2e6dd9c6b",
     "data/dependencies-test.parquet": "f5cd9430efaeb829bf84048dcfe34959f7400277973daa2c22b43b646938a6d1",
     "data/scenarios-test.parquet": "1e0c8aebfe2b749b3cadd8384a699ad8c3f47965146379c63391d5b863b29d6f",
     "data/tasks-test.parquet": "3d1f78c3d69281e6cf33e75d3b99e5adeecd3ba93315af6e0e5d65b902f2117b"