Query and analyze SEC filings and financial statements using EdgarTools. Get company data, filings, XBRL financials, and perform multi-company analysis.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Analyze SEC filings and financial statements using EdgarTools
Essential SEC filing analysis operations. See objects.md for object reference, workflows.md for patterns, readme.md for setup.
REQUIRED: Set your identity (SEC requirement):
from edgar import set_identity
set_identity("Your Name your@email.com")
Without this, all API calls fail with "User-Agent identity is not set" error.
ALWAYS use .to_context() first for concise summaries with available actions. 5-10x more token-efficient than full objects.
from edgar import Company
company = Company("AAPL")
print(company.to_context()) # ~88 tokens vs 200+ for full object
Output:
**Company:** Apple Inc.
**CIK:** 0000320193
**Ticker:** AAPL
**Exchange:** Nasdaq
**Industry:** Electronic Computers (SIC 3571)
**Fiscal Year End:** Sep 30
filings = company.get_filings(form="10-K")
print(filings.to_context()) # ~95 tokens vs 500-1000 for rich table
Shows summary + AVAILABLE ACTIONS.
filing = filings.latest()
print(filing.to_context()) # ~109 tokens, includes available methods
xbrl = filing.xbrl()
print(xbrl.to_context()) # ~275 tokens vs 2,500+ for full statements
Token Comparison:
| Object | Full Output | to_context() | Savings |
|---|---|---|---|
| Company | ~200 tokens | ~88 tokens | 56% |
| Filings | ~500-1000 | ~95 tokens | 80-90% |
| XBRL | ~2,500 tokens | ~275 tokens | 89% |
Pattern: to_context() first → see available → access data.
Common starting patterns. Use .to_context() for efficiency.
from edgar import set_identity, Company
set_identity("Your Name your@email.com") # Required first!
company = Company("AAPL")
print(company.to_context()) # Concise profile (~88 tokens)
# OR for full details:
# print(company) # Full object (~200 tokens)
from edgar import get_current_filings
filings = get_current_filings() # Last ~24 hours
print(filings.to_context()) # Summary + available actions (~95 tokens)
# OR to see first 5 in table:
# print(filings.head(5)) # Rich table (~500-1000 tokens)
from edgar import Company
company = Company("AAPL")
income = company.income_statement(periods=3) # 3 fiscal years
print(income) # Full statement
Main API functions and approaches.
Choose the approach based on your use case:
When to use: Cross-company screening, pattern discovery, historical research, don't know which specific companies.
Data source: SEC quarterly indexes (updated nightly)
from edgar import get_filings
# Get all filings for a quarter
filings = get_filings(2023, 1) # Q1 2023
# Filter by form type
filings = get_filings(2023, 1, form="10-K")
# Filter by date range
filings = get_filings(2023, 1, filing_date="2023-02-01:2023-02-15")
# Further filter results
filtered = filings.filter(ticker="AAPL")
tech_filings = filings.filter(ticker=["AAPL", "MSFT", "GOOGL"])
When to use: Monitoring recent filing activity, tracking latest submissions
Data source: SEC RSS feed (last ~24 hours)
from edgar import get_current_filings
# Get all recent filings
current = get_current_filings()
# Filter by form type
reports = current.filter(form=["10-K", "10-Q"])
# Filter by specific companies
tech_current = current.filter(ticker=["AAPL", "MSFT"])
When to use: You know the specific company ticker or name
Data source: SEC company submissions endpoint
from edgar import Company
company = Company("AAPL")
# Get all filings
all_filings = company.get_filings()
# Filter by form type
annual_reports = company.get_filings(form="10-K")
# Filter by year
filings_2023 = company.get_filings(year=2023)
# Combine filters
q1_2023_10q = company.get_filings(year=2023, form="10-Q")
When to use: Comparing multiple periods, trend analysis (fastest approach)
Data source: SEC Company Facts API
Advantages: Very fast (single API call), pre-aggregated data, multi-period comparison built-in
from edgar import Company
company = Company("AAPL")
# Annual data (fiscal years)
income = company.income_statement(periods=3) # Last 3 fiscal years
balance = company.balance_sheet(periods=3)
cash_flow = company.cash_flow_statement(periods=3)
# Quarterly data
quarterly_income = company.income_statement(periods=4, annual=False) # Last 4 quarters
When to use: Need specific filing details, want complete line items, analyzing single period
Data source: XBRL files attached to specific filings
Advantages: Most comprehensive detail, all line items available, exact as-filed data
from edgar import Company
company = Company("AAPL")
# Get specific filing
filing = company.get_filings(form="10-K")[0] # Latest 10-K
# Parse XBRL
xbrl = filing.xbrl()
# Get statements
income = xbrl.statements.income_statement()
balance = xbrl.statements.balance_sheet()
cash_flow = xbrl.statements.cash_flow_statement()
# Access metadata
print(f"Entity: {xbrl.entity_name}")
print(f"Fiscal Year: {xbrl.fiscal_year}")
print(f"Period: {xbrl.fiscal_period}")
⚠️ IMPORTANT: Filing has TWO different search methods. Use the right one!
filing.search(query) ⭐ Find Text in FilingsSearch the actual filing document - find keywords, topics, or sections within SEC filings.
from edgar import Company
company = Company("AAPL")
filing = company.get_filings(form="DEF 14A")[0] # Proxy statement
# Search for content IN the filing
results = filing.search("executive compensation")
# Process results
print(f"Found {len(results)} matches")
for match in results[:5]: # Top 5 matches
print(f"Relevance score: {match.score:.2f}")
print(f"Excerpt: {str(match)[:200]}...")
print()
Features: BM25 relevance ranking (best matches first), searches parsed HTML sections, returns DocSection objects with scores, index cached for performance (~1-2 seconds per filing)
Use cases: Find mentions of specific topics ("revenue recognition", "risk factors"), locate sections in large filings, screen filings for relevant content, extract context around keywords
Example: Find proxy statements mentioning compensation changes
from edgar import get_filings
from datetime import datetime, timedelta
# Get recent proxy statements
start_date = datetime.now() - timedelta(days=30)
filings = get_filings(form="DEF 14A")
recent = filings.filter(filing_date=f"{start_date.strftime('%Y-%m-%d')}:")
# Search each filing
companies_with_matches = []
for filing in recent:
matches = filing.search("executive compensation changes")
if matches and len(matches) > 0:
companies_with_matches.append({
'company': filing.company,
'date': filing.filing_date,
'matches': len(matches),
'top_score': matches[0].score,
'excerpt': str(matches[0])[:200]
})
print(f"Found {len(companies_with_matches)} companies")
filing.docs.search(query) 📚 Find MethodsSearch the Filing API documentation - discover how to use the Filing class.
# Find how to use Filing API
help_text = filing.docs.search("how to get XBRL")
print(help_text) # Shows documentation about filing.xbrl() method
help_text = filing.docs.search("convert to markdown")
print(help_text) # Shows documentation about filing.markdown() method
Use cases:
| What are you searching? | Method | Returns |
|---|---|---|
| Text in the filing (content) | filing.search("keyword") | List of DocSection matches with scores |
| How to use Filing API (methods) | filing.docs.search("how to") | API documentation snippets |
⚠️ Common Mistake:
# WRONG - Searches API docs, not filing content!
matches = filing.docs.search("executive compensation") # ❌
# Returns empty - API docs don't mention "executive compensation"
# CORRECT - Searches the actual filing document
matches = filing.search("executive compensation") # ✅
# Returns 50+ matches from proxy statement
Complete examples in common-questions.md.
| Task | Primary Method | Example |
|---|---|---|
| Show S-1 filings from date range | get_filings(year, quarter, form="S-1", filing_date="...") | See example |
| Get today's filings | get_current_filings() | See example |
| Get company revenue trend | company.income_statement(periods=3) | See example |
| Get quarterly financials | company.income_statement(periods=4, annual=False) | See example |
| Get statement from specific filing | filing.xbrl().statements.income_statement() | See example |
| Compare multiple companies | compare_companies_revenue(["AAPL", "MSFT"]) | See example |
| Get latest quarterly balance sheet | company.get_filings(form="10-Q")[0].xbrl() | See example |
| Get insider transactions (Form 4) | company.get_filings(form="4") | See example |
| Filter filings efficiently | filings.filter(ticker="AAPL", filing_date="2024-01-01:") | See example |
| Look up form types | describe_form("C") or see form-types-reference.md | See example |
Pattern: For any question, check common-questions.md for full working examples.
Advanced patterns, helpers, error handling, skill exportation: advanced-guide.md.
Includes:
Error:
RuntimeError: User-Agent identity is not set. Please call set_identity() first.
Cause: Missing set_identity() call (SEC requirement)
Solution:
from edgar import set_identity
set_identity("Your Name your@email.com") # Must call before any API operations
Error:
AttributeError: 'Company' object has no attribute 'sic_code'
Cause: Incorrect attribute name
Solution: Check the API reference in objects.md for correct attribute names (e.g., use company.sic instead of company.sic_code)
Cause: Not using .to_context() method
Solution: Always call .to_context() before printing full objects:
# Instead of:
print(company) # 200+ tokens
# Use:
print(company.to_context()) # ~88 tokens
Problem: get_filings() returns empty list
Possible causes: No filings match criteria (try broader search), wrong quarter/year combination, or form type doesn't exist for that period
Solution:
filings = get_filings(2024, 1, form="10-K")
if len(filings) == 0:
print("No filings found - try different criteria")
# Try broader search
all_filings = get_filings(2024, 1)
print(f"Found {len(all_filings)} total filings in 2024 Q1")
Use the skill API to read documentation:
from edgar.ai import get_skill
skill = get_skill("EdgarTools")
common_questions = skill.get_document_content("common-questions")
advanced_guide = skill.get_document_content("advanced-guide")
Available documents: SKILL, common-questions, advanced-guide, quickstart-by-task, objects, workflows, form-types-reference, readme
See readme.md for complete API documentation.
EdgarTools automatically handles SEC rate limiting (10 requests/second):
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Testing patterns for PHPUnit and Playwright E2E tests. Use when writing tests, debugging test failures, setting up test coverage, or implementing test patterns for ActivityPub features.
日本語の概要は準備中です。原文の説明を表示しています。
Cloud laboratory platform for automated protein testing and validation. Use when designing proteins and needing experimental validation including binding assays, expression testing, thermostability measurements, enzyme activity assays, or protein sequence optimization. Also use for submitting experiments via API, tracking experiment status, downloading results, optimizing protein sequences for better expression using computational tools (NetSolP, SoluProt, SolubleMPNN, ESM), or managing protein design workflows with wet-lab validation.
日本語の概要は準備中です。原文の説明を表示しています。
Add unsigned integer (uint) type support to PyTorch operators by updating AT_DISPATCH macros. Use when adding support for uint16, uint32, uint64 types to operators, kernels, or when user mentions enabling unsigned types, barebones unsigned types, or uint support.
日本語の概要は準備中です。原文の説明を表示しています。
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
日本語の概要は準備中です。原文の説明を表示しています。
Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid search, and distributed systems integration. Use when building distributed AI systems, multi-agent coordination, or advanced vector search applications.
日本語の概要は準備中です。原文の説明を表示しています。
Create and train AI learning plugins with AgentDB's 9 reinforcement learning algorithms. Includes Decision Transformer, Q-Learning, SARSA, Actor-Critic, and more. Use when building self-learning agents, implementing RL, or optimizing agent behavior through experience.
日本語の概要は準備中です。原文の説明を表示しています。