Skip to content

RECONSTRUCTION-PLAN.md

Latest commit

 

History

History
82 lines (66 loc) · 3.43 KB

RECONSTRUCTION-PLAN.md

File metadata and controls

82 lines (66 loc) · 3.43 KB
Raw

Reconstruction Plan

First Principles

What this app does: Check if gaming streamers are live. Generate briefings. That's it.

What it pretends to do: 4-agent distributed system with LLM intent classification, Prometheus metrics, 3 cache layers, resource optimization, connection pooling, rate limiting, concurrency control.

Ratio: 19,000 lines of code → ~500 lines of actual product logic → 38:1 bloat ratio.

What Gets Cut

Component Lines Why Cut
router_agent.py 1052 4-agent routing is fake distribution. Replace with 20-line intent function.
briefing_agent.py 1210 Multi-agent collaboration is calling functions in the same process.
data_source_agent.py 883 Wraps a simple API call in 883 lines of abstraction.
data_source_agent_cached.py 253 Duplicate of above with cache bolted on.
main.py 378 4-agent orchestrator.
main_enhanced.py 142 Duplicate main with different cache.
llm_client.py 537 LLM intent classification for 7 regex patterns.
error_handler.py 538 Recovery managers, cooldowns, severity enums for a demo.
performance_metrics.py 570 Prometheus metrics for a single-user demo.
cache_optimizer.py 503 3 cache layers (LRU, Query, DataSource) for mock data.
resource_optimizer.py 417 psutil monitoring, connection pools, rate limiters for a demo.
server_config.py 43 Merged into app.py.
router_enhanced.py 212 Dead code (imports non-existent functions).
studio_helper.py 438 Help text generator. Merged into app.py.
response_formatter.py 362 Emoji formatter. Merged into app.py.
common.py 510 Performance monitor, detailed logger, pydantic models. Keep minimal utils.
huya_api.py 160 Not actually used (mock data is used). Cut.
twitch_api.py 340 Not actually used (mock data is used). Cut.
data_sources.py 576 Abstraction layer over mock data. Cut.
tests/ (13 files) ~2000+ Test the old architecture. Cut.
docs/ (19 files) ~3000+ Documentation of the old architecture. Cut.
scripts/ (3 files) ~200 Setup scripts for old architecture. Cut.
Various root files ~1000 Deploy scripts, verify scripts, reports. Cut.

What Gets Built (~310 lines, 5 files)

app.py (~120 lines)

  • Gradio web UI
  • Simple intent detection (regex, 15 lines)
  • Query routing to data functions
  • Health check endpoint
  • Entry point

data.py (~100 lines)

  • Mock streamer data (from existing mock_data.py, trimmed)
  • get_live_streams(game=None, streamer=None) → list of dicts
  • get_streamer_status(name) → single streamer dict
  • No classes, no abstractions, just functions

config.py (~30 lines)

  • Load players.yaml
  • Load .env
  • Simple dict access

requirements.txt (~8 lines)

  • Only what's actually needed: gradio, aiohttp, pyyaml, python-dotenv, loguru

Dockerfile (~15 lines)

  • Slim, minimal, copies only what exists

Files Kept As-Is

  • config/players.yaml - real config data
  • .env.example - env template
  • .gitignore
  • README.md - will be trimmed

Execution Order

  1. Create src/data.py (mock data + fetch functions)
  2. Create src/config.py (config loader)
  3. Create src/app.py (web UI + intent routing)
  4. Update requirements.txt
  5. Update Dockerfile
  6. Delete all old source files
  7. Delete tests/, docs/, scripts/, root clutter
  8. Update README.md
  9. Git commit