Earnings Calendar

Overview

This skill retrieves upcoming earnings announcements for US stocks using the Financial Modeling Prep (FMP) API. It focuses on companies with significant market capitalization (mid-cap and above, over $2B) that are likely to impact market movements. The skill generates organized markdown reports showing which companies are reporting earnings over the next week, grouped by date and timing (before market open, after market close, or time not announced).

Key Features:

• - Uses FMP API for reliable, structured earnings data
• Filters by market cap (>$2B) to focus on market-moving companies
• Includes EPS and revenue estimates
• Multi-environment support (CLI, Desktop, Web)
• Flexible API key management
• Organized by date, timing, and market cap

Prerequisites

FMP API Key

This skill requires a Financial Modeling Prep API key.

Get Free API Key:

1. 1. Visit: https://site.financialmodelingprep.com/developer/docs
2. Sign up for free account
3. Receive API key immediately
4. Free tier: 250 API calls/day (sufficient for weekly earnings calendar)

API Key Setup by Environment:

Claude Code (CLI):
CODEBLOCK0

Claude Desktop:
Set environment variable in system or configure MCP server.

Claude Web:
API key will be requested during skill execution (stored only for current session).

Core Workflow

Step 1: Get Current Date and Calculate Target Week

CRITICAL: Always start by obtaining the accurate current date.

Retrieve the current date and time:

• - Use system date/time to get today's date
• Note: "Today's date" is provided in the environment ( tag)
• Calculate the target week: Next 7 days from current date

Date Range Calculation:
CODEBLOCK1

Why This Matters:

• - Earnings calendars are time-sensitive
• "Next week" must be calculated from the actual current date
• Provides accurate date range for API request

Format dates in YYYY-MM-DD for API compatibility.

Step 2: Load FMP API Guide

Before retrieving data, load the comprehensive FMP API guide:

CODEBLOCK2

This guide contains:

• - FMP API endpoint structure and parameters
• Authentication requirements
• Market cap filtering strategy (via Company Profile API)
• Earnings timing conventions (BMO, AMC, TAS)
• Response format and field descriptions
• Error handling strategies
• Best practices and optimization tips

Step 3: API Key Detection and Configuration

Detect API key availability based on environment.

Multi-Environment API Key Detection:

3.1 Check Environment Variable (CLI/Desktop)

CODEBLOCK3

If environment variable is set, proceed to Step 4.

3.2 Prompt User for API Key (Desktop/Web)

If environment variable not found, use AskUserQuestion tool:

Question Configuration:
CODEBLOCK4

3.2.1 If user chooses "No, get free key":

Provide instructions:
CODEBLOCK5

3.3 Request API Key Input

If user has API key, request input:

Prompt:
CODEBLOCK6

Store API key in session variable:
CODEBLOCK7

Confirm with user:
CODEBLOCK8

Step 4: Retrieve Earnings Data via FMP API

Use the Python script to fetch earnings data from FMP API.

Script Location:
CODEBLOCK9

Execution:

Option A: With Environment Variable (CLI):
CODEBLOCK10

Option B: With Session API Key (Desktop/Web):
CODEBLOCK11

Script Workflow (automatic):

1. 1. Validates API key and date parameters
2. Calls FMP Earnings Calendar API for date range
3. Fetches company profiles (market cap, sector, industry)
4. Filters companies with market cap >$2B
5. Normalizes timing (BMO/AMC/TAS)
6. Sorts by date → timing → market cap (descending)
7. Outputs JSON to stdout

Expected Output Format (JSON):
CODEBLOCK12

Save to file (recommended for use with report generator):
CODEBLOCK13

Or capture to variable:
CODEBLOCK14

Error Handling:

If script returns errors:

• - 401 Unauthorized: Invalid API key → Verify key or re-enter
• 429 Rate Limit: Exceeded 250 calls/day → Wait or upgrade plan
• Empty Result: No earnings in date range → Expand date range or note in report
• Connection Error: Network issue → Retry or use cached data if available

Step 5: Process and Organize Data

Once earnings data is retrieved (JSON format), process and organize it:

5.1 Parse JSON Data

Load JSON data from script output:
CODEBLOCK15

Or if saved to file:
CODEBLOCK16

5.2 Verify Data Structure

Confirm data includes required fields:

• - ✓ symbol
• ✓ companyName
• ✓ date
• ✓ timing (BMO/AMC/TAS)
• ✓ marketCap
• ✓ sector

5.3 Group by Date

Group all earnings announcements by date:

• - Sunday, [Full Date] (if applicable)
• Monday, [Full Date]
• Tuesday, [Full Date]
• Wednesday, [Full Date]
• Thursday, [Full Date]
• Friday, [Full Date]
• Saturday, [Full Date] (if applicable)

5.4 Sub-Group by Timing

Within each date, create three sub-sections:

1. 1. Before Market Open (BMO)
2. After Market Close (AMC)
3. Time Not Announced (TAS)

Data is already sorted by timing from the script, so maintain this order.

5.5 Within Each Timing Group

Companies are already sorted by market cap descending (script output):

• - Mega-cap (>$200B) first
• Large-cap ($10B-$200B) second
• Mid-cap ($2B-$10B) third

This prioritization ensures the most market-moving companies are listed first.

5.6 Calculate Summary Statistics

Compute:

• - Total Companies: Count of all companies in dataset
• Mega/Large Cap Count: Count where marketCap >= $10B
• Mid Cap Count: Count where marketCap between $2B and $10B
• Peak Day: Day of week with most earnings announcements
• Sector Distribution: Count by sector (Technology, Healthcare, Financial, etc.)
• Highest Market Cap Companies: Top 5 companies by market cap

Step 6: Generate Markdown Report

Use the report generation script to create a formatted markdown report from the JSON data.

Script Location:
CODEBLOCK17

Execution:

Option A: Output to stdout:
CODEBLOCK18

Option B: Save to file:
CODEBLOCK19

What the script does:

1. 1. Loads earnings data from JSON file
2. Groups by date and timing (BMO/AMC/TAS)
3. Sorts by market cap within each group
4. Calculates summary statistics
5. Generates formatted markdown report
6. Outputs to stdout or saves to file

The script automatically handles all formatting including:

• - Proper markdown table structure
• Date grouping and day names
• Market cap sorting
• EPS and revenue formatting
• Summary statistics calculation

Report Structure:

CODEBLOCK20

Formatting Best Practices:

• - Use markdown tables for clean presentation
• Bold important company names (mega-cap) if desired
• Include market cap in human-readable format ($3.0T, $150B, $5.2B) - already formatted by script
• Group logically by date then timing
• Include summary section at top for quick overview
• Add EPS and revenue estimates if available

Step 7: Quality Assurance

Before finalizing the report, verify:

Data Quality Checks:

1. 1. ✓ All dates fall within the target week (next 7 days)
2. ✓ Market cap values are present for all companies
3. ✓ Each company has timing specified (BMO/AMC/TAS)
4. ✓ Companies are sorted by market cap within each section
5. ✓ Summary statistics are accurate
6. ✓ Report generation date is clearly stated
7. ✓ EPS and revenue estimates included where available

Completeness Checks:

1. 1. ✓ All days of the target week are included (even if no earnings)
2. ✓ Major known companies are not missing (verify against external sources if needed)
3. ✓ Sector information is included where available
4. ✓ Timing reference section is present
5. ✓ Data sources are credited (FMP API)

Format Checks:

1. 1. ✓ Markdown tables are properly formatted
2. ✓ Dates are consistently formatted
3. ✓ Market caps use consistent units (B for billions, T for trillions)
4. ✓ All sections follow template structure
5. ✓ No placeholder text ([PLACEHOLDER]) remains
6. ✓ EPS and revenue estimates properly formatted

Step 8: Save and Deliver Report

Save the generated report with an appropriate filename:

Filename Convention:
CODEBLOCK21

Example: INLINECODE0

The filename date represents the report generation date, not the earnings week.

Delivery:

• - Save the markdown file to the working directory
• Inform the user that the report has been generated
• Provide a brief summary of key findings (e.g., "45 companies reporting next week, with Apple and Microsoft on Monday")

Example Summary:
CODEBLOCK22

Fallback Mode (Step 8 Alternative): Manual Data Entry

If API access is unavailable or user chooses to skip API:

Provide Instructions for Manual Entry:

CODEBLOCK23

Process Manual Input:

1. 1. Parse user-provided earnings data
2. Organize by date, timing, and market cap
3. Generate report using same template
4. Note in report: "Data Source: Manual Entry"

Use Cases and Examples

Use Case 1: Weekly Review (Primary Use Case)

User Request: "Get next week's earnings calendar"

Workflow:

1. 1. Get current date (e.g., November 2, 2025)
2. Calculate target week (November 3-9, 2025)
3. Load FMP API guide
4. Detect/request API key
5. Fetch earnings data:

   python scripts/fetch_earnings_fmp.py 2025-11-03 2025-11-09 > earnings_data.json
   

1. 6. Generate markdown report:

   python scripts/generate_report.py earnings_data.json earnings_calendar_2025-11-02.md
   

1. 7. Notify user with summary

Complete One-Liner:
CODEBLOCK26

Use Case 2: Focused on Specific Day

User Request: "What earnings are coming out Monday?"

Workflow:

1. 1. Get current date and identify next Monday (e.g., November 4, 2025)
2. Fetch full week data (same as Use Case 1)
3. Generate full report but highlight Monday section
4. Provide verbal summary of Monday's earnings with emphasis

Use Case 3: Mega-Cap Focus

User Request: "Show me earnings for companies over $100B market cap next week"

Workflow:

1. 1. Fetch full earnings data (script already filters >$2B)
2. Process and organize as normal
3. When generating report, add a "Mega-Cap Focus" section at top
4. Filter tables to show only companies >$100B
5. Note: Still include full data in appendix for reference

Use Case 4: Sector-Specific

User Request: "What tech companies have earnings next week?"

Workflow:

1. 1. Fetch full earnings data
2. Process and organize as normal
3. Filter results by sector = "Technology"
4. Generate report with focus on technology sector
5. Note: Template structure remains the same; content is filtered

Troubleshooting

Problem: API key not working

Solutions:

• - Verify API key is correct (copy-paste carefully)
• Check if API key is active (login to FMP dashboard)
• Ensure no extra spaces before/after key
• Try generating new API key from FMP dashboard

Problem: Script returns empty results

Solutions:

• - Verify date range is in future (not past dates)
• Check date format is YYYY-MM-DD
• Try wider date range (e.g., 14 days instead of 7)
• Verify companies actually have announced earnings dates for that week

Problem: Missing major companies

Solutions:

• - Company may not have announced earnings date yet
• Some companies announce dates very late (1-2 days before)
• Cross-reference with company investor relations website
• Market cap may have dropped below $2B threshold

Problem: Rate limit hit (429 error)

Solutions:

• - Free tier: 250 calls/day
• Each weekly report uses ~3-5 API calls
• Check if other tools/scripts are using same API key
• Wait 24 hours for rate limit reset
• Consider upgrading to paid tier if needed frequently

Problem: Script execution error

Solutions:

• - Verify Python 3 is installed: INLINECODE1
• Install requests library: INLINECODE2
• Check script has execute permissions: INLINECODE3
• Run with python3 explicitly: INLINECODE4

Best Practices

Do's

✓ Always get current date first before any data retrieval ✓ Use FMP API as primary source for reliability ✓ Store API key in environment variable for CLI usage ✓ Sort by market cap to prioritize high-impact companies ✓ Group by date then timing for logical organization ✓ Include summary statistics for quick overview ✓ Credit data sources in report footer ✓ Use clean markdown tables for readability ✓ Provide timing reference section for clarity ✓ Note data freshness and potential for changes ✓ Include EPS and revenue estimates when available

Don'ts

✗ Don't assume "next week" without calculating from current date ✗ Don't omit timing information (BMO/AMC/TAS) ✗ Don't mix date formats within report (stay consistent) ✗ Don't include micro/small-cap unless specifically requested ✗ Don't forget to sort by market cap within sections ✗ Don't share API key in conversations or reports ✗ Don't include earnings from current week or past dates ✗ Don't generate report without quality assurance checks ✗ Don't commit API keys to version control

Security Notes

API Key Security

Important Reminders:

1. 1. ✓ Use free tier API keys for testing
2. ✓ Rotate keys regularly
3. ✓ Don't share conversations containing API keys
4. ✓ Set API key as environment variable for CLI
5. ✓ Keys provided in chat are session-only (forgotten after session ends)
6. ✗ Never commit API keys to Git repositories
7. ✗ Never use production API keys with sensitive data access

Best Practice:
For Claude Code (CLI), always use environment variable:
CODEBLOCK27

For Claude Web, understand that:

• - API key entered in chat is temporary
• Stored only in conversation context
• Not saved to disk
• Forgotten when session ends

Resources

FMP API:

• - Main Documentation: https://site.financialmodelingprep.com/developer/docs
• Get API Key: https://site.financialmodelingprep.com/developer/docs
• Earnings Calendar API: https://site.financialmodelingprep.com/developer/docs/earnings-calendar-api
• Company Profile API: https://site.financialmodelingprep.com/developer/docs/companies-key-metrics-api
• Pricing/Rate Limits: https://site.financialmodelingprep.com/developer/docs/pricing

Supplementary Sources (for verification):

• - Seeking Alpha: https://seekingalpha.com/earnings/earnings-calendar
• Yahoo Finance: https://finance.yahoo.com/calendar/earnings
• MarketWatch: https://www.marketwatch.com/tools/earnings-calendar

Skill Resources:

• - FMP API Guide: INLINECODE5
• Python Script: INLINECODE6
• Report Template: INLINECODE7

---

Summary

This skill provides a reliable, API-driven approach to generating weekly earnings calendars for US stocks. By using FMP API, it ensures structured, accurate data with additional insights like EPS/revenue estimates. The multi-environment support makes it flexible for CLI, Desktop, and Web usage, while the fallback mode ensures functionality even without API access.

Key Workflow: Date Calculation → API Key Setup → API Data Retrieval → Processing → Report Generation → QA → Delivery

Output: Clean, organized markdown report with earnings grouped by date/timing/market cap, including summary statistics and trading considerations.