Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.b1e55ed.permanentupperclass.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Developer Guide

How to extend and contribute to b1e55ed.

​
Project Structure

b1e55ed/
├── api/                    # FastAPI app and routes
│   ├── main.py             # App factory, router mounts
│   ├── auth.py             # Bearer token auth
│   ├── deps.py             # FastAPI dependencies
│   └── routes/             # Route modules by domain
├── engine/                 # Core engine
│   ├── brain/              # Brain orchestration
│   ├── core/               # DB, config, events, webhooks
│   ├── integrations/       # EAS, forge, oracle
│   ├── producers/          # Signal producers
│   └── strategies/         # Backtest strategies
├── dashboard/              # Starlette dashboard
├── docs/                   # Documentation
├── tests/                  # Test suite
│   ├── unit/               # Fast, isolated tests
│   └── integration/        # End-to-end tests
├── config/                 # Config files
│   ├── default.yaml        # Repo defaults
│   └── user.yaml           # Operator overlay (gitignored)
└── scripts/                # Utilities and validation

​
Key Modules

ModulePurpose
engine/core/db.pySQLite event store, hash chain
engine/core/config.pyConfig model, env var mapping
engine/core/contributors.pyContributor registry, scoring
engine/core/scoring.pyAttribution scoring logic
engine/core/webhooks.pyWebhook dispatch
engine/brain/orchestrator.pyBrain cycle orchestration
engine/brain/synthesis.pyMulti-domain conviction synthesis
engine/brain/regime.pyRegime detection
engine/integrations/oracle.pyOracle provenance projection
engine/integrations/eas.pyEAS attestation client
engine/integrations/forge.pyIdentity derivation
api/routes/signals.pySignal submit, attribution
api/routes/mcp.pyMCP JSON-RPC server
api/routes/oracle.pyOracle endpoint
api/routes/events.pySSE event stream

​
Test Setup

uv run pytest --ignore=tests/unit/test_eas_client.py -x -q
The EAS client test is excluded by default — it requires external network access. For a specific test file: 
uv run pytest tests/unit/test_brain.py -v

​
Adding a Producer

Producers live in engine/producers/. They emit signal events to the event store.

​
1. Create the producer module

# engine/producers/your_producer.py

from engine.core.db import EventStore
from engine.core.events import emit_event

class YourProducer:
    """Generates signals from [your source]."""

    name = "your_producer"
    domain = "technical"  # or: onchain, social, tradfi, curator

    def __init__(self, config: dict):
        self.config = config

    def run(self, db: EventStore) -> list[str]:
        """Run one cycle. Return list of emitted event IDs."""
        event_ids = []

        # Fetch data, compute signal
        direction = "bullish"  # or: bearish, neutral
        conviction = 7.0

        event_id = emit_event(
            db=db,
            event_type=f"signal.{self.domain}.your_signal.v1",
            source=f"producer.{self.name}",
            payload={
                "symbol": "BTC",
                "direction": direction,
                "conviction": conviction,
                "rationale": "...",
            },
        )
        event_ids.append(event_id)
        return event_ids

​
2. Register via CLI

b1e55ed producers register \
  --name your_producer \
  --domain technical \
  --endpoint http://localhost:8001/poll \
  --schedule "*/15 * * * *"
Or via config in config/user.yaml for built-in producers.

​
3. Write tests

# tests/unit/test_your_producer.py

def test_your_producer_emits_signal(tmp_db):
    producer = YourProducer(config={})
    event_ids = producer.run(tmp_db)
    assert len(event_ids) > 0

​
Adding a Strategy

Strategies live in engine/strategies/. They consume signals and produce trade intents for the backtest engine. 
# engine/strategies/your_strategy.py

from engine.strategies.base import BaseStrategy

class YourStrategy(BaseStrategy):
    """Multi-factor strategy combining [signals]."""

    name = "your_strategy"

    def evaluate(self, signals: list[dict]) -> list[dict]:
        """Return list of trade intents."""
        intents = []
        # Group by symbol, check consensus, size via Kelly
        return intents
Register in the strategy registry used by the backtest CLI.

​
Extension Registry Pattern

The engine uses a registry pattern for producers and strategies. Built-in components are discovered at import time. External components register via the database or config overlay. For producers: engine/core/producers.py — ProducerRegistry For strategies: engine/strategies/__init__.py — StrategyRegistry

​
Contribution Guidelines

  1. Branch from develop — all PRs target develop, not main
  2. Tests required — new code without tests won’t merge
  3. No broken links — run bash scripts/validate_doc_deps.sh before commit
  4. Lint clean — uv run ruff check . && uv run ruff format --check .
  5. Type-check — uv run mypy engine/ api/ for changed modules

​
CI Requirements

Every PR must pass: 
uv run pytest --ignore=tests/unit/test_eas_client.py -x -q   # tests
uv run ruff check .                                            # lint
uv run ruff format --check .                                   # format
bash scripts/validate_doc_deps.sh                              # doc deps
Type checking runs on CI for the engine/ and api/ packages.

​
See Also