https://github.com/ai-riksarkivet/oxenstierna
🦬⭐ MCPs for Riksarkivet open APIs. This covers both agents and mcps
https://github.com/ai-riksarkivet/oxenstierna
agents iiif iiif-image iiif-presentation mcp mcp-server oai-pmh riksarkivet search-engine smolagents
Last synced: 3 months ago
JSON representation
🦬⭐ MCPs for Riksarkivet open APIs. This covers both agents and mcps
- Host: GitHub
- URL: https://github.com/ai-riksarkivet/oxenstierna
- Owner: AI-Riksarkivet
- License: mit
- Created: 2025-06-05T09:40:12.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-06-13T05:14:05.000Z (4 months ago)
- Last Synced: 2025-06-13T06:26:59.050Z (4 months ago)
- Topics: agents, iiif, iiif-image, iiif-presentation, mcp, mcp-server, oai-pmh, riksarkivet, search-engine, smolagents
- Language: Python
- Homepage:
- Size: 761 KB
- Stars: 3
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Oxenstierna 🦬⭐ (WIP)
## MCPs for Riksarkivet
Alot of mcps can be built with the help of: https://github.com/Riksarkivet/dataplattform/wiki
- Riksarkivet OAIPMH API Integration: https://github.com/Riksarkivet/dataplattform/wiki/OAI-PMH
- Riksarkivet IIIF API Integration: https://github.com/Riksarkivet/dataplattform/wiki/IIIF
- Riksarkivet Search API Integration https://github.com/Riksarkivet/dataplattform/wiki/Search-API
- (new) can we use this? with semantic search: https://forvaltningshistorik.riksarkivet.se/Index.htm <---
- AI-Riksarkviet HTRflow pypi
___
Current implementation:
```
This server provides access to the Swedish National Archives (Riksarkivet) through multiple APIs.
SEARCH-BASED WORKFLOW (start here):
- search_records: Search for content by keywords (e.g., "coffee", "medical records")
- get_collection_info: Explore what's available in a collection
- get_all_manifests_from_pid: Get all image batches from a collection
- get_manifest_info: Get details about a specific image batch
- get_manifest_image: Download specific images from a batch
- get_all_images_from_pid: Download all images from a collection
URL BUILDING TOOLS:
- build_image_url: Build IIIF Image URLs with custom parameters
- get_image_urls_from_manifest: Get all URLs from an image batch
- get_image_urls_from_pid: Get all URLs from a collection
TYPICAL WORKFLOW:
1. search_records("your keywords") → find PIDs
2. get_collection_info(pid) → see what's available
3. get_manifest_info(manifest_id) → explore specific image batch
4. get_manifest_image(manifest_id, image_index) → download specific image
Example PID: LmOmKigRrH6xqG3GjpvwY3
```
___
### Tools
- Riksarkivet OAIPMH Metadata API Integration
- Riksarkivet IIIF Image API Integration
- Riksarkivet IIIF Presentation API Integration
#### Search API
https://sok.riksarkivet.se/?Sokord=kaffe&f=True&EndastDigitaliserat=true&EndastDigitaliserat=false&TranskriberadText=true&TranskriberadText=false&Fritext=&Namn=&Ort=&DatumFran=&DatumTill=&AvanceradSok=false
The idea here is that you can use the search api to get hits across transcribed text:
https://data-acc.riksarkivet.se/api/records?transcribed_text=kaffe&only_digitised_materials=false&offset=0&limit=100&sort=relevance
IIIF content search is not a viable option!
#### HTRflow
Would it be possible to HTR on the flow?
https://huggingface.co/spaces/Gabriel/htrflow_mcp
schema: https://gabriel-htrflow-mcp.hf.space/gradio_api/mcp/schema
TODO:
- mcp https://github.com/knaw-huc/pagexml
- sperate content from outpit file
- visualizer
#### Resource
We want to resources simlair to this concept
- https://gradio-docs-mcp.hf.space/
Part of Riksarkivet OAIPMH API Integration can act as resource.
Also, iiif presentation and image api https://github.com/IIIF/api/blob/main/source/image/3.0/index.md
## FastMCP 2.0
https://gofastmcp.com/getting-started/welcome
## MCP Inspector Integration
- Use inspector for interactive testing and debugging
## Client
### Claude Desktop Integration
- Document how to add the server to Claude Desktop
### Aider
..