The v2 prompt produces expected_move for every ticker mention. This
commit makes KevinStrategy.evaluate_mention USE it as a hard signal
rather than just a display field.
Three new rules, all guarded by KevinStrategyConfig knobs so the
behaviour can be turned off if it over-filters:
1) SELL + non-bearish expected_move => NO_OP (require_forward_for_
bearish, default True). This is THE anti-capitulation rule —
Kevin saying "I sold" without articulating where the stock goes
next becomes NO_OP. Reactive sells stop translating into
trades.
2) AVOID + bullish expected_move => NO_OP (don't close, don't
blocklist). Same idea — if the LLM's forward call contradicts the
avoid action, treat as inconsistent and skip.
3) BUY + bearish/sideways expected_move => NO_OP (schema veto).
Catches LLM inconsistency.
4) BUY + unknown expected_move => bump min_conviction floor by
unknown_conviction_bonus (default +0.05). Forces stronger
conviction when there's no forward direction.
Tests: 6 new (one per rule above), 22 regression — total 28 GREEN.
Backtest stub _mention factory now defaults expected_move from
action (buy/sell/avoid maps) so existing backtest scenarios stay
green; the test_backtest_sell_mid_position_closes_early case was
the only one that needed the fix.
Side note: strategy is backward-compatible. If a mention has no
expected_move attribute (e.g. v1 stub from older code), it defaults
to UNKNOWN and the legacy code paths still work — just with the
stricter conviction floor on buys.
User reported that the old prompt could emit 'sell' on backward-looking
capitulation ('Kevin sold after a 20% drop') — exactly the false signal
to avoid. v2 reframes every per-ticker field as forward-looking and
adds an explicit expected_move enum for the trading bot to weight.
Changes:
- New ExpectedMove enum (up_strong/up_mild/sideways/down_mild/
down_strong/unknown) in shared/schemas + shared/models, with
matching kevin_expected_move Postgres enum + column on
kevin_stock_mentions (migration e5f6a7b8c9d0). NOT NULL with
server_default 'unknown' so existing rows backfill cleanly.
- SYSTEM_PROMPT rewritten: action semantics now require a FORWARD
view; reactive sells get downgraded to 'watch' or skipped; the
rationale_quote must contain forward reasoning. Quality
checklist updated.
- _ANALYSIS_TOOL JSON schema gains expected_move (required).
- prompt_version v1 → v2 in config + infra + ad-hoc CLI default.
- pipeline.py persists ticker.expected_move into the new column.
Migration safety: the column is NOT NULL DEFAULT 'unknown' so 96
existing mentions auto-fill with 'unknown' (no forward call known
for backward analyses) without breaking any reads.
Cost to backfill the 27 existing analyses with v2 prompt: ~$3 LLM
spend. A follow-up reanalyze script will replay them after this
ships.
Wires the dependency-injected KevinBridge to concrete Redis cursor +
DB session factory + AlpacaBroker (or stub when creds missing). Includes
TradeSignalPublisher (Pydantic -> dict for the redis stream) and
SIGINT/SIGTERM graceful shutdown. Adds is_asset_tradable + get_latest_price
to AlpacaBroker so the bridge can query asset metadata.
Composable: cursor/aggregator/strategy/publisher/audit_writer/broker
all injected. Master kill-switch (kevin_enable_trading=false) routes to
audit-only path. Cursor advances ONLY after XADD succeeds (race fix).
Concrete collaborators wired in subsequent tasks.
Also extends TradeSignal + SignalDirection.EXIT with the optional
fields Kevin paths need (strategy_id, target_dollars, stop_loss_pct,
take_profit_pct).
Stateless: mention + account_state -> KevinDecision. Conviction-weighted
sizing, time_horizon-derived hold periods, hard per-ticker cap. The
bridge and the backtest mini-engine both call evaluate_mention so
behaviour cannot drift.
3 tables (kevin_signal_bridge_state, kevin_backtest_runs,
kevin_backtest_trades) all UUID-keyed for consistency with Trade/Position.
KEVIN_STRATEGY_UUID constant pinned for FK joins from Trade.strategy_id.
Three issues caught during end-to-end manual QA against docker-compose:
1. SAEnum field columns serialized to Python enum NAMES ('DISCOVERED')
but the DB enum had VALUES ('discovered'). Added `values_callable`
to all 5 SAEnum() declarations in shared/models/meet_kevin.py so they
emit values, matching the migration's enum literals.
2. /dashboard's "last 7 days" / "last 14 days" filters used
`func.cast("7 days", type_=None)` which produced NullType DDL.
Replaced with `text("now() - interval '7 days'")`.
3. /dashboard's outlook trend query repeated `func.date_trunc("day", col)`
in SELECT, GROUP BY and ORDER BY — Postgres treats each as a separate
parameterized expression. Hoisted into a single `day_trunc` variable
so all three clauses reference the same SQL fragment.
All 11 /api/meet-kevin/* endpoints now return valid JSON against a
docker-compose Postgres seeded with one analyzed video + NVDA mention.
- Replace all Literal[...] type annotations with corresponding enum classes
(TickerAction, TimeHorizon, MarketOutlook, VideoStatus, TranscriptSource)
for MeetKevinTickerMention, MeetKevinAnalysis, and API response models
(VideoSummary, VideoDetail, StockMention, StockSummary, TimelineBucket)
- Add min_length=1, max_length=10 validation to MeetKevinTickerMention.symbol
- Split test_conviction_edge_cases into two separate boundary tests
- Strengthen test_valid_ticker_mention with assertions for all 6 fields
- Trim no-information docstrings from TranscriptSegment, StockTimeline
- All 60 schema tests pass
- Learning engine: expand default weights from 3 to all 9 strategies
- Learning engine: resolve placeholder strategy_id with DB lookup
- Learning engine: pass strategy_sources from trade execution
- Trade executor: respect trading:paused Redis flag in RiskManager
- Portfolio sync: compute actual daily P&L from day-start snapshot
- Portfolio API: cumulative P&L from first snapshot, read pause flag
- Portfolio metrics: compute max drawdown and avg hold duration
- Add strategy_sources field to TradeExecution schema
- Add dev_mode config (TRADING_DEV_MODE) to bypass auth for local dev
- Dashboard: VITE_DEV_MODE bypasses ProtectedRoute and 401 redirects
- Vite proxy target configurable via VITE_API_TARGET
- Add top-level README.md and remaining-work-plan.md
- Update CLAUDE.md with correct counts and remove stale TODOs
- 404 tests passing
Made-with: Cursor
- Point Ollama to local instance via host.docker.internal, use gemma3 model
- Remove Docker Ollama service (using host's Ollama instead)
- Add company-name-to-ticker mapping (Apple→AAPL, Tesla→TSLA, etc.) for RSS articles
- Lower signal thresholds for faster feedback with paper trading:
- FinBERT confidence: 0.6→0.4, signal strength: 0.3→0.15
- News strategy: article_count 2→1, confidence 0.5→0.3, score ±0.3→±0.15
- Fix market data BarSet access bug (BarSet.__contains__ returns False incorrectly)
- Fix market data SIP feed error by switching to IEX feed for free Alpaca accounts
- Fix nginx proxy routing for /api/auth/* to api-gateway /auth/*
- Add seed_sample_data script
- Update tests for new thresholds and alpaca mock modules
Wire the trading bot to real Alpaca market data and persist pipeline
state to the database so the dashboard displays live information.
- Add market-data service fetching OHLCV bars from Alpaca, publishing
to market:bars Redis Stream; signal generator consumes bars and
injects current_price into signals for position sizing
- Sentiment analyzer now persists Article + ArticleSentiment rows to
DB after scoring, with duplicate and error handling
- API gateway runs a background portfolio sync task that snapshots
Alpaca account state into PortfolioSnapshot/Position DB tables
during market hours
- TradeSignal carries a signal_id UUID; signal generator and trade
executor both persist their records to DB with cross-references
- 303 unit tests pass (57 new tests added)
I1: Add graceful shutdown (SIGTERM/SIGINT) to all 5 background services
I2: Fix Dockerfile healthcheck to use curl on /metrics endpoint
I3: Fix StreamConsumer.ensure_group() to only catch BUSYGROUP errors
I4: Fix SimulatedBroker to reject orders with insufficient cash/shares
I5: Move ORM attribute access inside DB session context in trades routes
I6: Add Redis-based rate limiting (10 req/min/IP) on all auth endpoints
I8: Prevent backtest background task garbage collection
I9: Use Numeric(16,6) instead of Float for financial columns in migration
I10: Add index on trades.created_at for time-range queries
I11: Bind infrastructure ports to 127.0.0.1 in docker-compose
I12: Add migrations init service; all app services depend on it
I13: Fix user enumeration in login_begin (return options for non-existent users)
- pyproject.toml with core deps and optional dep groups per service
- shared/config.py: Pydantic BaseSettings with TRADING_ env prefix
- shared/redis_streams.py: StreamPublisher/StreamConsumer wrappers
- shared/telemetry.py: OpenTelemetry + Prometheus metric export
- tests for Redis Streams helpers (5 passing)
