chore: establish standards in AGENTS.md

This commit is contained in:
Justin Visser 2026-08-09 20:56:13 +02:00
commit adcf4b9df6
4 changed files with 141 additions and 0 deletions

17
.gitignore vendored Normal file
View file

@ -0,0 +1,17 @@
# secrets
.env
# python
__pycache__/
*.pyc
.venv/
.mypy_cache/
.ruff_cache/
.pytest_cache/
# node
node_modules/
frontend/dist/
# eval: raw (unredacted) recordings never enter the repo
eval/fixtures/raw/

82
AGENTS.md Normal file
View file

@ -0,0 +1,82 @@
# Engineering standards
This file is the contract for every contributor to this repository — human or
agent. `CLAUDE.md` is a symlink to it: one canonical standard, read by every
tool. `ruff`, `mypy`, `eslint`, `prettier`, and `vue-tsc` enforce mechanically
what they can; this document carries the rest.
## Architecture
Monorepo with a ports-and-adapters seam. The dependency direction is enforced
by review:
```
api → pipeline → ports ← adapters domain imports nothing app-level
```
- `backend/app/domain/` — pure models and logic (`Track`, `TasteProfile`,
fuzzy matching, profile compression). Imports nothing app-level.
- `backend/app/ports/``Protocol` definitions (`MusicCatalog`,
`Recommender`, `PlaylistWriter`).
- `backend/app/adapters/` — implementations: `spotify/` (auth, thin httpx
client, DTO mapping), `anthropic/` (both LLM calls), `demo/`
(fixture-replay implementations of the same ports).
- `backend/app/pipeline/` — the orchestrator composes the stages; stages never
import each other.
- Spotify JSON never escapes `adapters/spotify/mapping.py`.
- Pipeline tunables (counts, thresholds, budgets, model id, effort, TTLs) live
in `app/config.py` (pydantic-settings) — no magic numbers in code.
- LLM prompts are versioned template files under `app/prompts/`, not inline
strings.
- `eval/scenarios.yaml` is one source of truth for three consumers: golden
eval queries, demo fixture keys, and UI suggestion chips.
## Naming
Human-readable, descriptive, full words. Code reads as prose.
- Python: `snake_case` functions/variables, `PascalCase` classes,
`UPPER_SNAKE` constants. Functions verb-first (`resolve_candidate`),
variables noun-first (`grounded_tracks`), booleans as predicates
(`is_grounded`, `should_stream`).
- TypeScript/Vue: `camelCase` variables/functions, `PascalCase` components and
`.vue` filenames, composables `use`-prefixed, types without `I`-prefix.
- Abbreviations: only universally idiomatic ones (`id`, `url`, `api`, `db`,
`llm`, `sse`). Never invented ones.
- Single-letter names only in lambdas/comprehensions whose whole scope is
≤ ~3 lines.
- Modules named for what they contain, singular nouns. `utils.py`,
`helpers.py`, `misc.py` are banned names.
## Sizing, DRY, modularity
- Soft cap ~300 lines per module, hard cap 500 — split by responsibility.
Functions target ≤ ~40 lines.
- One responsibility per module: if it needs "and" to describe, split it.
- DRY by the rule of three: extract on the third occurrence, not the second.
No premature abstraction, no helpers for one-shot operations, no designing
for hypothetical future requirements.
- Validate at boundaries only (user input, Spotify responses, LLM output);
trust internal code — no defensive re-checking between our own modules.
## Comments & docstrings
Sparse and load-bearing:
- One-line docstring on every module and public function/class; add
Args/Returns only where the signature genuinely doesn't tell the story.
- Inline comments state only what the code cannot — constraints, API quirks,
non-obvious whys. Never narrate the next line.
- Banned: commented-out code, drive-by TODOs, decorative section banners.
- Type hints everywhere; `mypy --strict` is the enforcement.
## Enforcement
- Backend: `ruff` (lint + format) everywhere. `mypy --strict` on `domain/`,
`ports/`, `pipeline/`; default strictness on `adapters/` and `api/`
SDK/streaming typing archaeology is not where the effort goes. Every
`# type: ignore` carries an error code and a reason.
- Frontend: `eslint` + `prettier` + `vue-tsc --noEmit`; no `any`. The stream
event types are a discriminated union mirrored against the Pydantic
response schemas.
- Commit messages describe the change, conventional-commit style.

1
CLAUDE.md Symbolic link
View file

@ -0,0 +1 @@
AGENTS.md

41
README.md Normal file
View file

@ -0,0 +1,41 @@
# discovery-by-llm
Muziekontdekking met een LLM als aanbevelingsmotor, op de Spotify Web API.
**These:** Spotify heeft zijn aanbevelings- en audio-intelligentielaag uit de
API verwijderd (recommendations en audio features zijn weg voor nieuwe apps).
Deze proof-of-concept beantwoordt de vraag of een LLM die rol kan overnemen,
met Spotify teruggebracht tot resolver, verifier, personalisatiebron en
afspeeloppervlak. Een naïeve-zoekbaseline maakt het antwoord meetbaar.
> Werk in uitvoering — dit bestand wordt gevuld tijdens de bouw.
> Chronologisch verslag: [docs/logboek.md](docs/logboek.md).
## Demo
*(volgt: gehoste instantie + lokale `docker compose up --build`, met en zonder
API-sleutels)*
## Hoe het werkt
*(volgt: pipeline-diagram en module-overzicht)*
## Keuzes
*(volgt)*
## Performance en optimalisaties
*(volgt)*
## Waar te kijken
*(volgt)*
## Werkwijze
*(volgt)*
## Wat ik heb geschrapt / wat ik hierna zou doen
*(volgt)*