---
name: caselaw-access-project-ingestion
description: "How to get Harvard Caselaw Access Project data into a pandas DataFrame and Parquet: access, the request, the format, limits, quirks and terms."
---

# Load CAP reporter metadata into pandas and Parquet

Source entry: https://signal-lab-jet.vercel.app/datasets/caselaw-access-project
Provider: https://static.case.law/
Checked against the provider's documentation: 2026-11-01

## Access
- No account or key. Static JSON metadata, volume ZIPs, case HTML and PDFs are public downloads.
- This example uses the small root reporter-metadata file; full case text is partitioned by reporter and volume.

## Request
```bash
curl -O 'https://static.case.law/ReportersMetadata.json'
```

## Format
JSON list of reporters with IDs, names, year range, jurisdiction objects, Harvard catalog IDs and reporter slug.

## Load into pandas and Parquet
Needs Python 3 with requests, pandas and pyarrow.
```python
"""CAP reporter metadata -> pandas -> Parquet. No key."""
import json
import requests
import pandas as pd

URL = "https://static.case.law/ReportersMetadata.json"
r = requests.get(URL, timeout=60)
r.raise_for_status()
df = pd.DataFrame(r.json())
df["jurisdictions"] = df["jurisdictions"].map(json.dumps)
df["harvard_hollis_id"] = df["harvard_hollis_id"].map(json.dumps)
df.to_parquet("cap_reporters_metadata.parquet", index=False)
print(len(df), "reporters", int(df["start_year"].min()), "to",
      int(df["end_year"].max()))
```
Run by us on 2026-11-01 (python 3.13.5 | pandas 3.1.0 | pyarrow 26.0.0 | requests 2.34.2). Result: 401 reporters 1658 to 2020

## Limits
- Rate limit: Not stated.
- Coverage ends with volumes published in 2020 and is not kept current.
- Full bulk volume files can be large; the checked root VolumesMetadata.json was about 30 MB.

## Quirks
- Reporter and jurisdiction fields are nested JSON; serialize or normalize them before writing a flat Parquet table.
- Case text is raw OCR and can contain recognition errors even where key metadata was reviewed.
- PDFs with selectable OCR are offered through 2018, while structured case data extends through 2020.
- The collection omits many unpublished, trial-document and digital-only decisions.

## Terms and attribution
- Caselaw data and metadata are marked CC0 1.0 Universal and provided as is.
- Harvard requests attribution and freely available improvements as nonbinding community norms.
- Terms page: https://case.law/terms/

## Sources
- https://case.law/templates/cap-docs-page.js: "Our static file export provides access to the following file types:"
- https://case.law/templates/cap-about-page.js: "The CAP data is free for the public to use and access."
- https://case.law/templates/cap-terms-page.js: "Fellows of Harvard College is marked with CC0 1.0 Universal"
- https://case.law/templates/cap-about-page.js: "Our earliest case is from 1658, and we currently include all volumes"

## Help improve this skill
Missing a source, or see something to improve? Call the MCP tool suggest_source or feedback (or POST /api/suggest or /api/feedback). Reviewed ideas: https://signal-lab-jet.vercel.app/agent-suggestions

Information for educational purposes only. Free tiers, limits and licenses belong to the providers and may change; check the provider's own terms before using the data.