Reported issues for dannet

Pod holds 6 of 6 problems reported by people outside the maintainer team. Issues filed by the project's own owners, members and collaborators are excluded entirely — a maintainer's release checklist is not a warning to a prospective user.

Back to dannet.

Most discussed

MCP server (MVP)

Using this issue to plot the possible development of an MCP server targeting DanNet.

Current challenges

• I probably need a subscription in order to develop the solution as it'll require a fair bit of testing.
  • Claude costs $20/month which is a fairly substantial cost.
• DanNet itself will have to be modified with a (relatively open) API
  • ... unless the MCP server is local-only, e.g. it spins up a local RDF graph and queries this graph.

Outstanding issues

• Ontological t

Read the thread · 2025-08-20 · closed · 13 comments

Compatibility with goodmami/wn

I'd like to be able to query DanNet through github.com/goodmami/wn.

As far as I can see all it requires is an endpoint with an WN-LMF 1.1 XML file available somewhere on a stable url.

Read the thread · 2024-05-19 · closed · external user · 12 comments

Include similarity functionality

e.g. similar to the ones in the wn tool: https://wn.readthedocs.io/en/latest/api/wn.similarity.html

This could be a pre-generated stat, but also a dynamic endpoint taking two RDF resources and the specific algorithm apply.

Read the thread · 2026-04-09 · closed · 3 comments

Remote MCP server

The DanNet MCP server should also be available as a remote MCP server. In fact, this should probably be the default way to access the DanNet MCP server.

However, the locally running version (the one initially developed in #155) should still exist and should probably be the exact same Python codebase. I can probably put the MCP server behind Caddy just like the DanNet web service. Local development would happen using t

Read the thread · 2025-09-03 · closed · 2 comments

Performance improvements to SPARQL and MCP server access

Currently, most of the more advanced query results of the MCP server rely on SPARQL queries which often aren't very performant and will therefore time out or just never return any data. For this reason, we need to

• improve the performance of SPARQL queries
  • setting a bigger timeout
  • suggesting query templates that won't be as performance intensive
• document alternate ways of finding information in the MCP server
  • i.e. fixed queries for various common issues, perhaps with caching and

Read the thread · 2026-03-04 · closed · 1 comment

JSON-LD export broken

Since shifting from sets to sets + seqs, the JSON-LD export has been broken.

Read the thread · 2025-10-06 · closed · 0 comments

Most recent

The remaining reports are on the project's issue tracker.