AI Monorepo
AI-agnostisches, versioniertes Wissens-Repo als Single Source of Truth für beliebige AIs. Heterogene Eingaben (YouTube, URLs, PDF, Audio, Bild, Slides) werden automatisch zu sauberem Markdown aufbereitet und über einen MCP-Server für jede MCP-fähige AI abrufbar gemacht.
Konzept & Entscheidungen: siehe
AI-Monorepo-Konzept.md(im Begleit-RepoMacOS-monitoring).
Zwei Repos — bewusst getrennt
| Repo | Inhalt |
|---|---|
dieses (ai-monorepo) |
Code: Ingestion, Index, MCP-Server, (später) App |
| Store (separat, privat) | Daten = Source of Truth: Markdown + Frontmatter, vom Tooling beschrieben |
Der Store-Pfad wird über die Config (monorepo.example.toml → monorepo.toml) gesetzt. Der
Code committet nie sich selbst in den Store.
Struktur
src/monorepo/
pipeline/ Ingestion-Orchestrierung
adapters/ je Quellentyp ein Adapter (youtube, pdf, url, audio, …)
index/ Vektorindex (LanceDB), aus dem Store gebaut
mcp/ MCP-Server (search_context …)
cli.py Kommandozeilen-Einstieg
docs/
frontmatter-schema.md Vertrag für jeden Store-Eintrag
Leitprinzipien
- Embeddings/Index = wegwerfbarer Cache, nie Source of Truth (→ rebuild statt migrieren).
- Originale werden zu Pointern; das Transkript ist die SoT.
- Pipeline-AI veredelt beim Ablegen, Roh-Original bleibt immer erhalten.
- MCP filtert zur Abfragezeit nach Lebenszyklus (
as_of-Zeitreise möglich). - Lokal-first, „nur ich".
Phasen-Status
- Phase 0 — Skelett & Konventionen
- Phase 1 — YouTube-Adapter
- Phase 2 — AI-Veredelung
- Phase 3 — Index (LanceDB)
- Phase 4 — MCP-Server
- Phase 5 — Lebenszyklus + weitere Adapter
- Phase 6 — Host + Mac-Fallback
- Phase 7 — App + CaptureKit (eigener Track)
Entwicklung
python3 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
ruff check .
Metadata
Release files for ai-monorepo 0.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| ai_monorepo-0.2.0.tar.gz | 34.1 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ai_monorepo-0.2.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 71.4 kB
Release files / ai_monorepo-0.2.0.tar.gz
| Download URL | ai_monorepo-0.2.0.tar.gz |
|---|---|
| Size | 34.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
487b5efb5ec340b6a6d341a97ba2955c3ffdc0a69a414e0e0db5321e84fbae06
|
|
BLAKE2b-256 checksum How to use checksums |
13335c2543ee30190460e92504edc93f0d1838cb01d4902a9a38fe600cdbbc3b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.5
|
Release files / ai_monorepo-0.2.0-py3-none-any.whl
| Download URL | ai_monorepo-0.2.0-py3-none-any.whl |
|---|---|
| Size | 37.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e70c123b28980431536aade7e581184a133ab1b9e2465d299df0d69ab871f2b1
|
|
BLAKE2b-256 checksum How to use checksums |
4efd3c65b7b6246ac910e0564ba4617d824a937e505720e81fd066e0da5f0e1d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.5
|
Security for ai-monorepo 0.2.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.