Add small agent runtime telemetry dataset

README.md

@@ -0,0 +1,104 @@
+---
+pretty_name: Agent Runtime Telemetry Small
+license: other
+language:
+  - en
+tags:
+  - agent-runtime
+  - agent-observability
+  - llm-observability
+  - mcp
+  - tool-calling
+  - runtime-telemetry
+  - audit-trail
+  - workflow-traces
+  - parquet
+size_categories:
+  - 10K<n<100K
+configs:
+  - config_name: dataset_overview
+    data_files:
+      - split: train
+        path: data/dataset_overview.parquet
+  - config_name: operations
+    data_files:
+      - split: train
+        path: data/operations.parquet
+  - config_name: operation_events
+    data_files:
+      - split: train
+        path: data/operation_events.parquet
+  - config_name: artifact_records
+    data_files:
+      - split: train
+        path: data/artifact_records.parquet
+  - config_name: audit_records
+    data_files:
+      - split: train
+        path: data/audit_records.parquet
+  - config_name: tool_summary
+    data_files:
+      - split: train
+        path: data/tool_summary.parquet
+  - config_name: artifact_summary
+    data_files:
+      - split: train
+        path: data/artifact_summary.parquet
+  - config_name: daily_activity
+    data_files:
+      - split: train
+        path: data/daily_activity.parquet
+---
+
+# Agent Runtime Telemetry Small
+
+Agent Runtime Telemetry Small is a compact tabular export of MCP-style agent execution telemetry. It is designed for dataset viewer inspection, lightweight agent observability experiments, tool-call reliability analysis, workflow trace summaries, and audit-trail research.
+
+The dataset is intentionally small and row-oriented. Each table is stored as Parquet so the Hugging Face Dataset Viewer can display clean columns without requiring a SQLite client.
+
+## What It Contains
+
+| Config | Rows | Columns | Purpose |
+|---|---:|---:|---|
+| `dataset_overview` | 7 | 6 | Table inventory and export policy |
+| `operations` | 2,262 | 33 | Tool execution records, status, stages, durations, and summarized result metadata |
+| `operation_events` | 9,903 | 13 | Lifecycle events for operations |
+| `artifact_records` | 1,269 | 19 | Forecast, state-decode, and training artifact index records |
+| `audit_records` | 14,053 | 17 | Tool request/result audit rows with compact metadata |
+| `tool_summary` | 32 | 8 | Aggregated tool reliability and latency statistics |
+| `artifact_summary` | 9 | 7 | Aggregated artifact status and payload-size statistics |
+| `daily_activity` | 8 | 5 | UTC daily activity counts across runtime tables |
+
+## Privacy Boundary
+
+This export does not upload the original SQLite databases and does not include raw nested `payload_json` bodies. Large JSON fields are represented with inspectable columns such as key lists, byte lengths, selected scalar status fields, and SHA-256 digests. Absolute local paths are reduced to path scope and file name columns.
+
+## Suggested Uses
+
+- compare agent tool success/error rates across runtime traces
+- inspect workflow latency and stage transitions
+- prototype LLM agent observability dashboards
+- analyze audit request/result volume without parsing full JSON logs
+- benchmark small-data telemetry pipelines that expect clean tabular inputs
+
+## Loading Example
+
+```python
+from datasets import load_dataset
+
+ops = load_dataset("YOUR_USERNAME/agent-runtime-telemetry-small", "operations")
+print(ops["train"][0])
+
+summary = load_dataset("YOUR_USERNAME/agent-runtime-telemetry-small", "tool_summary")
+print(summary["train"].to_pandas().sort_values("operation_count", ascending=False).head())
+```
+
+## Source
+
+The rows were exported from local runtime SQLite stores into sanitized Parquet tables:
+
+- `operation_state.sqlite3`
+- `artifact_store.sqlite3`
+- `audit_store.sqlite3`
+
+The export focuses on the operational shape of agent runtimes rather than application-specific content.

UPLOAD_INSTRUCTIONS.md

@@ -0,0 +1,93 @@
+# Upload Instructions
+
+Target dataset name:
+
+```text
+agent-runtime-telemetry-small
+```
+
+Recommended repo id:
+
+```text
+YOUR_HF_USERNAME/agent-runtime-telemetry-small
+```
+
+## Files To Upload
+
+Upload the full contents of this folder:
+
+```text
+data/huggingface_exports/agent-runtime-telemetry-small/
+```
+
+Do not upload the original runtime SQLite files. This folder already contains the viewer-friendly Parquet export and the Hugging Face dataset card.
+
+## CLI Upload
+
+From the repository root:
+
+```bash
+python3 - <<'PY'
+from pathlib import Path
+import os
+
+from dotenv import load_dotenv
+from huggingface_hub import HfApi, create_repo, upload_folder
+
+root = Path("data/huggingface_exports/agent-runtime-telemetry-small").resolve()
+load_dotenv(".env")
+
+token = (
+    os.getenv("HF_TOKEN")
+    or os.getenv("HUGGINGFACE_HUB_TOKEN")
+    or os.getenv("HUGGING_FACE_HUB_TOKEN")
+)
+if not token:
+    raise SystemExit("Missing HF_TOKEN or HUGGINGFACE_HUB_TOKEN.")
+
+api = HfApi(token=token)
+username = api.whoami(token=token)["name"]
+repo_id = f"{username}/agent-runtime-telemetry-small"
+
+create_repo(repo_id=repo_id, repo_type="dataset", token=token, exist_ok=True, private=False)
+upload_folder(
+    repo_id=repo_id,
+    repo_type="dataset",
+    folder_path=str(root),
+    token=token,
+    commit_message="Add small agent runtime telemetry dataset",
+)
+
+print(f"https://huggingface.co/datasets/{repo_id}")
+PY
+```
+
+## Manual Web Upload
+
+1. Create a new Hugging Face Dataset named `agent-runtime-telemetry-small`.
+2. Upload `README.md`, `export_manifest.json`, and the full `data/` directory from this folder.
+3. Wait for the Dataset Viewer to process the Parquet files.
+4. Confirm the configs appear as `operations`, `operation_events`, `artifact_records`, `audit_records`, `tool_summary`, `artifact_summary`, `daily_activity`, and `dataset_overview`.
+
+## Validation After Upload
+
+```python
+from datasets import load_dataset
+
+repo_id = "YOUR_HF_USERNAME/agent-runtime-telemetry-small"
+for config in ["operations", "operation_events", "artifact_records", "audit_records", "tool_summary"]:
+    ds = load_dataset(repo_id, config)
+    print(config, ds["train"].num_rows, ds["train"].column_names[:8])
+```
+
+## Export Policy
+
+This folder is a sanitized export:
+
+- no source SQLite database files
+- no raw nested `payload_json` bodies
+- no absolute local paths
+- no secret-like token strings
+- Parquet tables optimized for Hugging Face Dataset Viewer
+
+If you regenerate from newer runtime state, keep the same policy so the dataset remains useful and safe to browse.

data/artifact_records.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1e02f24eb11596dd12ca2e59f4be88b24872b1727c1b96bdc8463b268c1dd92
+size 103040

data/artifact_summary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7e98cf8a1de3ad840e8e04a275ac1cf9454e29278e1e08d9e72465e812d1017
+size 5806

data/audit_records.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44c223bf1fdb2817ce461c2cbf650f31723fe1a7008d41d7686535eff87f5f9d
+size 2247227

data/daily_activity.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:616e42658d086cf811ed2b72673154748842d33f4bd78cbe30cc61bc4195877e
+size 3827

data/dataset_overview.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e8cb25a66e296926875fc6d16bd9fd9ff3659ddf3566e3036e63495be2529c5
+size 4796

data/operation_events.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4db53085f861baff539e12c1a238dc82ec840d881f7b6ca02c488f3d70e96584
+size 610446

data/operations.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fb8a1b19d5ae1f439fa8c3a09a45e8965801fd85937ef424788467dc7700655
+size 276568

data/tool_summary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb012f2611ac6d4d45bcffdac21c13d785f7298129ea9bd288b65d7603529f58
+size 8030

export_manifest.json

@@ -0,0 +1,200 @@
+{
+  "name": "agent-runtime-telemetry-small",
+  "created_at_utc": "2026-04-21T15:04:52.808915+00:00",
+  "source_runtime_databases": {
+    "operation_state": {
+      "path": "data/runtime_state/operation_state.sqlite3",
+      "size_bytes": 201281536
+    },
+    "artifact_store": {
+      "path": "data/runtime_state/artifact_store.sqlite3",
+      "size_bytes": 43900928
+    },
+    "audit_store": {
+      "path": "data/runtime_state/audit_store.sqlite3",
+      "size_bytes": 191569920
+    }
+  },
+  "export_root": "data/huggingface_exports/agent-runtime-telemetry-small",
+  "tables": {
+    "operations": {
+      "rows": 2262,
+      "columns": [
+        "operation_id",
+        "request_id",
+        "tool_name",
+        "status",
+        "stage",
+        "duration_ms",
+        "created_at_utc",
+        "updated_at_utc",
+        "args_fingerprint",
+        "args_count",
+        "args_keys",
+        "kwargs_key_count",
+        "kwargs_keys",
+        "operation_mode",
+        "backend_preference",
+        "force_retrain",
+        "include_control_sensitivities",
+        "include_validation_protocols",
+        "has_input_provenance",
+        "has_source_binding",
+        "series_rows_count",
+        "scenario_rows_count",
+        "result_summary_key_count",
+        "result_summary_keys",
+        "result_type",
+        "result_operation",
+        "result_payload_key_count",
+        "result_payload_keys",
+        "result_payload_bytes",
+        "artifacts_bytes",
+        "error_type",
+        "error_message_preview",
+        "error_message_sha256"
+      ],
+      "column_count": 33,
+      "file": "data/operations.parquet",
+      "size_bytes": 276568
+    },
+    "operation_events": {
+      "rows": 9903,
+      "columns": [
+        "event_id",
+        "operation_id",
+        "event_type",
+        "status",
+        "stage",
+        "event_time_utc",
+        "payload_bytes",
+        "payload_sha256",
+        "payload_key_count",
+        "payload_keys",
+        "payload_status",
+        "payload_stage",
+        "payload_tool"
+      ],
+      "column_count": 13,
+      "file": "data/operation_events.parquet",
+      "size_bytes": 610446
+    },
+    "artifact_records": {
+      "rows": 1269,
+      "columns": [
+        "record_id",
+        "artifact_kind",
+        "artifact_file",
+        "artifact_path_scope",
+        "config_tag",
+        "schema_name",
+        "status",
+        "payload_sha256",
+        "metadata_key_count",
+        "metadata_keys",
+        "metadata_status",
+        "metadata_source_mode",
+        "metadata_state_count",
+        "payload_key_count",
+        "payload_keys",
+        "payload_bytes",
+        "has_forecast",
+        "state_count",
+        "recorded_at_utc"
+      ],
+      "column_count": 19,
+      "file": "data/artifact_records.parquet",
+      "size_bytes": 103040
+    },
+    "audit_records": {
+      "rows": 14053,
+      "columns": [
+        "record_id",
+        "category",
+        "record_name",
+        "record_file",
+        "record_path_scope",
+        "tool",
+        "kind",
+        "status",
+        "duration_ms",
+        "request_id",
+        "payload_bytes",
+        "payload_sha256",
+        "payload_key_count",
+        "payload_keys",
+        "response_key_count",
+        "response_keys",
+        "created_at_utc"
+      ],
+      "column_count": 17,
+      "file": "data/audit_records.parquet",
+      "size_bytes": 2247227
+    },
+    "tool_summary": {
+      "rows": 32,
+      "columns": [
+        "tool_name",
+        "status",
+        "operation_count",
+        "avg_duration_ms",
+        "median_duration_ms",
+        "p95_duration_ms",
+        "first_seen_utc",
+        "last_seen_utc"
+      ],
+      "column_count": 8,
+      "file": "data/tool_summary.parquet",
+      "size_bytes": 8030
+    },
+    "artifact_summary": {
+      "rows": 9,
+      "columns": [
+        "artifact_kind",
+        "status",
+        "artifact_count",
+        "avg_payload_bytes",
+        "median_payload_bytes",
+        "first_recorded_utc",
+        "last_recorded_utc"
+      ],
+      "column_count": 7,
+      "file": "data/artifact_summary.parquet",
+      "size_bytes": 5806
+    },
+    "daily_activity": {
+      "rows": 8,
+      "columns": [
+        "date_utc",
+        "operations",
+        "operation_events",
+        "artifact_records",
+        "audit_records"
+      ],
+      "column_count": 5,
+      "file": "data/daily_activity.parquet",
+      "size_bytes": 3827
+    },
+    "dataset_overview": {
+      "rows": 7,
+      "columns": [
+        "table_name",
+        "row_count",
+        "column_count",
+        "file",
+        "viewer_role",
+        "payload_policy"
+      ],
+      "column_count": 6,
+      "file": "data/dataset_overview.parquet",
+      "size_bytes": 4796
+    }
+  },
+  "privacy_boundary": {
+    "raw_payload_json_uploaded": false,
+    "absolute_paths_uploaded": false,
+    "sqlite_databases_uploaded": false,
+    "secret_like_text_redaction": true,
+    "notes": "Export contains operational telemetry summaries derived from local runtime SQLite tables; large nested JSON payloads are represented by keys, byte lengths, selected scalar statuses, and SHA-256 digests."
+  }
+}