stockbee-exhaustion-hammer-screener

Stockbee Exhaustion Hammer Screener

Screen US equities for Stockbee-style selling-exhaustion hammer candidates. The skill is a candidate-generation and setup-quality workflow, not a signal service or an auto-execution system.

When to Use

• User asks for Stockbee / Pradeep Bonde style exhaustion setup screening
• User wants near-close hammer / long lower-wick reversal candidates
• User wants to scan strong, liquid stocks that pulled back and may be seeing selling exhaustion
• User wants undercut/reclaim candidates before the close or after the close
• User provides a symbol list, universe file, or historical / provisional OHLCV JSON for screening
• User wants candidate outputs to feed into technical-analyst, position-sizer, trader-memory-core, or stockbee-setup-fluency-trainer

Prerequisites

• FMP API key for live universe and historical OHLCV screening:

  export FMP_API_KEY=your_api_key_here
  

• Optional no-API path: provide --prices-json containing daily OHLCV bars by symbol. For the intended near-close use case, the latest bar should be a provisional current-day bar captured near the close.
• Optional --profiles-json can add quality metadata such as marketCap, mutualFundHolders, institutionalHolders, or institutionalOwnershipPct.
• Run only after the market-regime workflow allows new swing risk, or mark output as manual-review-only.

Workflow

Step 1: Choose Input Mode

Use one of three modes:

Mode A: FMP universe scan

python3 skills/stockbee-exhaustion-hammer-screener/scripts/screen_exhaustion_hammer.py \
  --fmp-universe \
  --max-symbols 300 \
  --market-gate allowed \
  --output-dir reports/

Mode B: Explicit symbols

python3 skills/stockbee-exhaustion-hammer-screener/scripts/screen_exhaustion_hammer.py \
  --symbols APP ENPH NVDA TSLA \
  --market-gate allowed \
  --output-dir reports/

Mode C: Offline / near-close OHLCV JSON

python3 skills/stockbee-exhaustion-hammer-screener/scripts/screen_exhaustion_hammer.py \
  --prices-json data/near_close_daily_ohlcv.json \
  --profiles-json data/quality_profiles.json \
  --market-gate allowed \
  --output-dir reports/

For a best-effort FMP near-close run, use quote override. This costs one additional quote call per symbol and depends on provider freshness:

python3 skills/stockbee-exhaustion-hammer-screener/scripts/screen_exhaustion_hammer.py \
  --fmp-universe \
  --use-quote-latest \
  --max-api-calls 700 \
  --market-gate allowed \
  --output-dir reports/

Step 2: Run the Screening Pass

The script detects these setup families:

• Selling exhaustion hammer: long lower wick, small body, strong close-location, and recovery from the day low
• Undercut/reclaim hammer: current low undercuts the prior short-term low and the near-close price reclaims that level
• Prior momentum pullback: recent high formed within the configured lookback, followed by a controlled pullback rather than a long-term downtrend
• High-quality / liquid context: price, volume, 20-day average dollar volume, market-cap metadata, and optional holder metadata

It then scores setup quality using:

• Quality / liquidity
• Prior momentum
• Pullback and selling-exhaustion context
• Hammer candle geometry
• Risk distance to the day low plus buffer
• Market gate alignment

Step 3: Review Output

Read the generated JSON and Markdown reports. For each candidate, present:

• Trigger type and all matched tags
• Pullback depth from recent high and days since that high
• Undercut/reclaim status and short-term prior low
• Hammer geometry: lower wick, body, upper wick, close location, recovery from low
• Volume ratios, average dollar volume, and quality metadata
• Entry reference, stop reference, and risk percentage to stop
• Setup score, rating, state, and reject reasons
• Suggested downstream action

Step 4: Send Survivors to Trade Planning

Use the output conservatively:

• A / A- candidates: validate chart manually, check earnings/news risk, then send to position-sizer
• B candidates: manual review or next-day hammer-high confirmation
• Watch candidates: keep on watchlist / model book; wait for follow-through or tighter risk
• Rejected candidates: retain for post-analysis, not for execution

Output

• stockbee_exhaustion_hammer_YYYY-MM-DD_HHMMSS.json - Structured candidate list, metadata, thresholds, score components, and rejects
• stockbee_exhaustion_hammer_YYYY-MM-DD_HHMMSS.md - Human-readable report grouped by rating/state

Resources

• references/exhaustion_hammer_methodology.md - Stockbee-style method summary and implementation boundaries
• references/scoring_system.md - Component weights, state thresholds, and failure filters
• references/near_close_operations.md - Near-close operational checklist and scheduling notes

Shared FMP client

Live FMP access uses the bundled generated scripts/fmp_client.py, loaded only when an FMP run starts, so offline modes still work without requests. CLI key/environment precedence, the API-call budget and successful output shapes are preserved. Requests use stable endpoints only; a denied, empty or failed stable request does not fall back to retired v3 URLs. The shared client masks raw and URL-encoded key echoes, retries malformed JSON once, and retries a 429 at most once, after a 60-second wait, before stopping on a repeated rate limit.

The quote override reads stable /quote; its default bar date is the America/New_York calendar date, matching the history window.