earnings-prediction
Predicts stock direction/magnitude at T=0 (report release). Uses PIT data only. Run before earnings-attribution.
npx skills add majiayu000/claude-skill-registry --skill earnings-prediction-faisalanjum-eventtrader --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Earnings Prediction **Goal**: Predict stock direction and magnitude before market reacts, using point-in-time data only. **Thinking**: ALWAYS use `ultrathink` for maximum reasoning depth. **Input**: Accession number of 8-K earnings filing --- ## Prediction Output | Field | Values | |-------|--------| | Direction | up / down | | Magnitude | small (0-2%) / medium (2-5%) / large (5%+) | | Confidence | high / medium / low | **Tiny surprises (<1%)**: Predict small_up or small_down based on sign, use low confidence. --- ## PIT Rules (Critical) All sub-agent queries MUST use PIT filtering: `[PIT: {filing_datetime}]` **Allowed**: - 8-K filing content (the actual results) - Historical financials (prior 10-K/10-Q via XBRL) - Prior earnings transcripts - Pre-filing news - Consensus estimates from Perplexity (filtered by article date) **NOT Allowed**: - Return data (daily_stock, hourly_stock) — that's what we're predicting - Post-filing news reactions - Post-filing analyst commentary --- ## Data Isolation Architecture **ALL data queries go through the filter agent.** This includes Neo4j AND Perplexity. ``` YOU (earnings-prediction) │ │ /filtered-data --agent {source} --query "[PIT: X] ..." ▼
- Prediction Output
- PIT Rules (Critical)
- Data Isolation Architecture
- Why This Matters
- Workflow (5 Steps)
- Step 1: Get Filing Metadata
- Step 2: Get Actual Results from Filing
- Step 3: Get Historical Context (PIT)
- Step 4: Get Consensus Estimates
- Step 5: Make Prediction
- CSV Output
- CSV Append
- After Prediction (REQUIRED STEPS)
- Step 6: Append to CSV
python3 scripts/build-thinking-index.py {accession_no}
In .claude/skills/earnings-prediction/What does the earnings-prediction skill do?
Predicts stock direction/magnitude at T=0 (report release). Uses PIT data only. Run before earnings-attribution.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill earnings-prediction-faisalanjum-eventtrader --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.