Add memnex schema

[UladzKha]

Summary

Adds memnex — an open specification for portable meeting outputs (transcripts, summaries, action items, decisions) from local-first meeting processing tools.

About memnex

memnex is a published JSON Schema for structured meeting data. It serves as an interop format between meeting processing pipelines (Whisper transcription + LLM analysis) and downstream consumers (note-taking apps, knowledge bases, AI agents via MCP).

• Spec repo: https://github.com/UladzKha/memnex
• Reference implementation: memnex-spec on npm
• Conformance suite: 15 test cases (7 valid, 8 invalid) — included in the spec repo
• License: Dual — MIT (code) + CC0 1.0 (spec)
• First adopter: Samuraizer (local-first meeting CLI)

Schema details

• JSON Schema draft: 2020-12
• fileMatch:
  • *.memnex.json — suffix convention for spec-aware tooling
  • meeting-output.json — default filename used by the reference implementation

Notes for reviewers

• The schema uses JSON Schema Draft 2020-12. I see Schemastore now supports it (Amend JSON Schema Draft 2020-12 definition #1977 was merged) — happy to discuss if draft-07 is preferred for this entry.
• The URL points to the main branch of the spec repo. A custom domain (memnex.org) is owned and will host the schema directly in a future iteration; once GitHub Pages is configured with the CNAME, I'll open a follow-up PR to update the URL.
• No tests included in this initial PR. The conformance suite lives in the spec repo and runs against the schema directly. I can submit Schemastore test cases as a follow-up if useful.

Thanks for maintaining Schemastore!

[github-actions]

Thanks for the PR!

This section of the codebase is owned by @madskristensen and @hyperupcall - if they write a comment saying "LGTM" then it will be merged.