DebtStack.ai

The credit API for AI agents

Corporate structure and debt analysis is complex. Even with AI, achieving accuracy, speed, and cost-effectiveness requires significant engineering. DebtStack.ai does this hard work once, giving you instant API access to pre-computed, quality-assured credit data.

Why DebtStack?

The Problem: Extracting accurate corporate structure and debt data from SEC filings is surprisingly hard:

Accuracy challenges: LLMs return malformed JSON, misinterpret amounts (cents vs dollars), confuse entity names, and aggregate data instead of extracting individual instruments
Speed variability: A single extraction can take 90-300 seconds with multiple LLM calls, retries, and QA loops
Cost uncertainty: Ad-hoc extraction costs $0.03-0.50+ per company, compounding with retries
Expertise required: Understanding 10-K structure, Exhibit 21, debt footnotes, VIEs, and credit agreement terminology

The Solution: DebtStack runs extraction once with rigorous QA, then serves pre-computed data via fast API.

Quick Start

1. Get Your API Key

Sign up at debtstack.ai or via the API:

curl -X POST "https://api.debtstack.ai/v1/auth/signup" \
  -H "Content-Type: application/json" \
  -d '{"email": "you@example.com", "password": "your-password"}'

2. Install the SDK (or use curl)

pip install debtstack-ai

3. Query

Python SDK:

from debtstack import DebtStackClient
import asyncio

async def main():
    async with DebtStackClient(api_key="ds_your_api_key") as client:
        # Screen for high-leverage companies
        risky = await client.search_companies(
            sector="Telecommunications",
            min_leverage=4.0,
            fields="ticker,name,net_leverage_ratio",
            sort="-net_leverage_ratio"
        )
        print(risky)

asyncio.run(main())

curl:

export DEBTSTACK_API_KEY="ds_your_api_key_here"

# Search companies with field selection
curl -H "X-API-Key: $DEBTSTACK_API_KEY" \
  "https://api.debtstack.ai/v1/companies?ticker=AAPL,MSFT,GOOGL&fields=ticker,name,net_leverage_ratio"

# Search bonds with pricing
curl -H "X-API-Key: $DEBTSTACK_API_KEY" \
  "https://api.debtstack.ai/v1/bonds?seniority=senior_unsecured&min_ytm=8.0"

# Traverse entity relationships (find guarantors)
curl -H "X-API-Key: $DEBTSTACK_API_KEY" \
  -X POST "https://api.debtstack.ai/v1/entities/traverse" \
  -H "Content-Type: application/json" \
  -d '{"start":{"type":"bond","id":"893830AK8"},"relationships":["guarantees"]}'

Current Database

314 companies covering S&P 100, NASDAQ 100, and beyond — 5,838 active debt instruments, 3,064 with real-time TRACE bond pricing, and searchable SEC filing sections.

Coverage spans all major sectors:

Sector	Sample Companies
Tech	AAPL, MSFT, NVDA, GOOGL, META, AMZN, TSLA, ORCL, AVGO
Financials	JPM, GS, BAC, WFC, MS
Healthcare	JNJ, UNH, LLY, ABBV, MRK, IDXX, DXCM
Consumer	WMT, COST, HD, MCD, KO, PEP, ORLY
Telecom/Cable	CHTR, ATUS, TMUS, LUMN
Energy	XOM, CVX, OXY
Offshore Drilling	RIG, VAL, DO, NE
Airlines	AAL, UAL, DAL
Semiconductors	NVDA, AMD, AVGO, GFS, CDNS

Features

Primitives API: 12 core endpoints optimized for AI agents with field selection and filtering
Three-Tier Pricing: Pay-as-You-Go ($0 + per-call), Pro ($199/mo unlimited), Business ($499/mo full access)
Individual Debt Instruments: Each bond, note, and credit facility extracted separately (not just totals)
Structured Covenants: 1,181 covenants (financial, negative, protective) linked to specific instruments
Guarantee Relationships: 3,831 guarantee records linking debt to guarantor subsidiaries
Collateral Tracking: 626 collateral records with asset types (equipment, vehicles, real estate, etc.)
Corporate Ownership Hierarchy: Nested parent-child structures from SEC Exhibit 21 and indenture parsing
Financial Statements: Quarterly income statement, balance sheet, cash flow from 10-Q/10-K
Credit Ratios: Leverage, interest coverage, margins, liquidity metrics
Bond Pricing: YTM and spread-to-treasury calculations (Finnhub/FINRA TRACE)
Treasury Yield History: 5+ years of daily US Treasury yields (1M-30Y tenors)
Document Search: Full-text search across 4,957 indentures and 2,946 credit agreements
Pre-computed Responses: Sub-second API serving via cached JSON with ETag support

API Endpoints

Authentication

Endpoint	Description
`POST /v1/auth/signup`	Create account and get API key
`GET /v1/auth/me`	Get current user info and credits (requires API key)

All other endpoints require an API key passed via X-API-Key header.

Primitives API (Optimized for AI Agents)

These 12 endpoints are designed for agents writing code - simple REST, field selection, powerful filtering.

Endpoint	Cost (Pay-as-You-Go)	Description
`GET /v1/companies`	$0.05	Search companies with field selection and 15+ filters
`GET /v1/bonds`	$0.05	Search/screen bonds with pricing, filters for yield, seniority, maturity
`GET /v1/bonds/resolve`	$0.05	Map bond identifiers - free-text to CUSIP (e.g., "RIG 8% 2027")
`GET /v1/financials`	$0.05	Quarterly financial statements (income, balance sheet, cash flow)
`GET /v1/collateral`	$0.05	Collateral securing debt (types, values, priority)
`GET /v1/covenants`	$0.05	Search structured covenant data (financial, negative, protective)
`GET /v1/covenants/compare`	Business only	Compare covenants across multiple companies
`GET /v1/companies/{ticker}/changes`	$0.10	Diff against historical snapshots
`POST /v1/entities/traverse`	$0.15	Graph traversal for guarantor chains, org structure
`GET /v1/documents/search`	$0.15	Full-text search across SEC filings
`POST /v1/batch`	Sum of ops	Execute multiple primitives in parallel
`POST /v1/coverage/request`	Free	Request coverage for non-covered companies

Example - Field Selection:

curl "/v1/companies?ticker=AAPL,MSFT&fields=ticker,name,net_leverage_ratio&sort=-net_leverage_ratio"

Example - Bond Search (Screening):

# Find high-yield senior unsecured bonds with pricing data
curl "/v1/bonds?seniority=senior_unsecured&min_ytm=8.0&has_pricing=true"

Example - Bond Resolve (Identifier Lookup):

# Map trader shorthand to CUSIP
curl "/v1/bonds/resolve?q=RIG%208%25%202027"

# Lookup by CUSIP
curl "/v1/bonds/resolve?cusip=89157VAG8"

Example - Financials (TTM):

# Get trailing twelve months financials for AAPL
curl "/v1/financials?ticker=AAPL&period=TTM"

# Get all Q3 2025 financials for comparison
curl "/v1/financials?fiscal_year=2025&fiscal_quarter=3&format=csv"

Example - Collateral (Recovery Analysis):

# Find all first-lien collateral for RIG
curl "/v1/collateral?ticker=RIG&priority=first_lien"

# Get collateral with valuations
curl "/v1/collateral?has_valuation=true&collateral_type=equipment"

Example - Entity Traversal:

curl -X POST "/v1/entities/traverse" -d '{"start":{"type":"bond","id":"893830AK8"},"relationships":["guarantees"]}'

Example - Document Search (Deep Dive on a Bond):

# After selecting a bond, ask questions about covenants, defaults, etc.
curl "/v1/documents/search?q=event+of+default&ticker=RIG&section_type=indenture"

Advanced Primitives

Traverse Entities - `POST /v1/entities/traverse`

Navigate the corporate structure graph. Follow guarantor chains, find subsidiary hierarchies, or trace ownership relationships.

Request Body:

{
  "start": {
    "type": "bond" | "entity" | "company",
    "id": "CUSIP, entity UUID, or ticker"
  },
  "relationships": ["guarantees", "subsidiaries", "parent"],
  "direction": "outbound" | "inbound" | "both",
  "max_depth": 3
}

Relationship Types:

Relationship	Direction	Description
`guarantees`	inbound	Find entities that guarantee a bond
`subsidiaries`	outbound	Find entities owned by a parent
`parent`	outbound	Find the parent of an entity
`issuer`	outbound	Find who issued a bond

Example - Find Bond Guarantors:

curl -X POST "/v1/entities/traverse" \
  -H "Content-Type: application/json" \
  -d '{
    "start": {"type": "bond", "id": "893830AK8"},
    "relationships": ["guarantees"],
    "direction": "inbound"
  }'

# Returns:
{
  "start": {
    "type": "bond",
    "id": "893830AK8",
    "name": "8.000% Senior Secured Notes due 2027"
  },
  "paths": [
    {
      "relationship": "guarantees",
      "entities": [
        {"name": "Transocean Inc.", "jurisdiction": "Cayman Islands"},
        {"name": "Transocean Offshore Deepwater Drilling Inc.", "jurisdiction": "Delaware"},
        {"name": "Triton Holding Company", "jurisdiction": "Cayman Islands"}
      ]
    }
  ],
  "total_guarantors": 47
}

Example - Find Corporate Hierarchy:

curl -X POST "/v1/entities/traverse" \
  -d '{
    "start": {"type": "company", "id": "CHTR"},
    "relationships": ["subsidiaries"],
    "max_depth": 2
  }'

Use Cases:

Structural subordination analysis (which subs guarantee which debt)
Recovery analysis (trace collateral through entity chains)
Regulatory exposure (find entities in specific jurisdictions)

Search Documents - `GET /v1/documents/search`

Full-text search across 13,862 SEC filing sections including indentures, credit agreements, and debt footnotes.

Parameters:

Parameter	Type	Description
`q`	string	Search query (supports phrases: `"event of default"`)
`ticker`	string	Filter to specific company
`section_type`	string	`indenture`, `credit_agreement`, `debt_footnote`, `exhibit_21`
`filing_type`	string	`10-K`, `10-Q`, `8-K`
`limit`	int	Max results (default: 10, max: 50)
`highlight`	bool	Include matching snippets (default: true)

Example - Find Covenant Language:

curl "/v1/documents/search?q=restricted+payment+dividend&ticker=RIG&section_type=indenture"

# Returns:
{
  "query": "restricted payment dividend",
  "total_matches": 12,
  "results": [
    {
      "section_type": "indenture",
      "document_title": "Indenture dated April 11, 2024",
      "filing_date": "2024-04-15",
      "snippet": "...shall not, and shall not permit any Restricted Subsidiary to, directly or indirectly, declare or pay any **dividend** or make any **payment** or distribution...",
      "relevance_score": 0.89,
      "source_url": "https://www.sec.gov/Archives/edgar/data/..."
    }
  ]
}

Common Search Queries:

Query	Documents	Use Case
`"event of default"`	3,608	Default triggers, grace periods
`"change of control"`	2,050	Put provisions, 101% repurchase rights
`collateral`	1,752	Security package analysis
`"asset sale"`	976	Mandatory prepayment triggers
`"make-whole"`	679	Early redemption premiums
`"restricted payment"`	464	Dividend/buyback restrictions

Use Cases:

Answer specific covenant questions ("Can they pay dividends?")
Find default triggers and grace periods
Research change of control provisions
Analyze collateral and security packages

Agent Workflow: Discovery → Deep Dive

DISCOVERY (Structured Data)              DEEP DIVE (Document Search)
───────────────────────────              ───────────────────────────
1. GET /v1/bonds?min_ytm=800
2. Filter by collateral, seniority
3. User picks specific bond ───────────► 4. GET /v1/documents/search
                                            ?q=covenant&ticker=XXX
                                         5. Agent summarizes snippets
                                         6. User sees plain English

DebtStack provides: Structured data + document snippets + source links Your agent provides: Query conversion + summarization + presentation

See docs/PRIMITIVES_API_SPEC.md for full specification.

System

Endpoint	Description
`GET /v1/ping`	Simple health check
`GET /v1/health`	Full health check with database
`GET /v1/status`	API status and data coverage
`GET /v1/sectors`	Sectors with company counts

Python SDK

Install the SDK for programmatic access or AI agent integration:

pip install debtstack-ai              # Core SDK
pip install debtstack-ai[langchain]   # LangChain tools (7 tools for credit analysis agents)
pip install debtstack-ai[mcp]         # MCP server (8 tools for Claude Desktop/Code/Cursor)

PyPI: pypi.org/project/debtstack-ai

See sdk/README.md for full documentation including LangChain agent setup and MCP client configurations.

MCP + LangChain

DebtStack integrates with AI agent frameworks out of the box:

LangChain: 7 tools for building credit analysis agents with any LLM. pip install debtstack-ai[langchain]
MCP Server: 8 tools for Claude Desktop, Claude Code, and Cursor. pip install debtstack-ai[mcp]

MCP Server Directories:

Directory	Link
Anthropic MCP Registry	registry.modelcontextprotocol.io
Smithery	smithery.ai
MCP Endpoint	mcp.debtstack.ai

See sdk/README.md for setup instructions, tool lists, and example queries.

Pricing

Tier	Price	Rate Limit	Includes
Pay-as-You-Go	$0/month	60/min	Pay per API call ($0.05-$0.15)
Pro	$199/month	100/min	Unlimited queries
Business	$499/month	500/min	Full access + historical pricing + bulk export + 5 team seats

Sign up at debtstack.ai or POST /v1/auth/signup.

Data Quality

Estimated Data Transparency

Estimated data is always flagged. When data cannot be extracted from SEC filings after repeated attempts and must be estimated or inferred, the API clearly indicates this:

issue_date_estimated: true - Issue date was inferred from maturity date and typical bond tenors (e.g., 10-year for senior notes), not extracted from the filing
Future estimated fields will follow the same pattern: {field}_estimated: true

This transparency ensures you always know when you're working with extracted data vs. inferred data.

Leverage Ratio Data Quality

Leverage ratios require TTM (Trailing Twelve Months) EBITDA calculations. We track data quality metadata so you know exactly how reliable each ratio is:

# Get leverage with data quality metadata
curl "/v1/companies?ticker=AAPL&include_metadata=true"

Returns:

{
  "ticker": "AAPL",
  "leverage_ratio": 0.63,
  "_metadata": {
    "leverage_data_quality": {
      "ebitda_source": "annual_10k",     // Used 10-K annual figures
      "ebitda_quarters": 4,               // Equivalent to 4 quarters
      "is_annualized": false,             // Not extrapolated
      "ebitda_estimated": false,          // D&A was available
      "ttm_quarters": ["2025FY"]          // Period used
    }
  }
}

TTM EBITDA Calculation Rules:

10-K filing: Use annual figures directly (already represents full year)
10-Q filing: Sum trailing 4 quarters
<4 quarters available: Annualize (flagged as is_annualized: true)
No D&A data: Use operating income as proxy (flagged as ebitda_estimated: true)

Ownership Relationship Transparency

Corporate ownership hierarchies are complex. SEC Exhibit 21 lists subsidiaries but rarely shows intermediate holding structures. We only show parent-child relationships where we have evidence:

{
  "structure": {
    "name": "Transocean Ltd.",
    "ownership_confidence": "root",
    "children": [
      {
        "name": "Transocean Inc.",
        "ownership_confidence": "key_entity",
        "children": [
          {
            "name": "Transocean Offshore Deepwater Holdings",
            "ownership_confidence": "verified"
          }
        ]
      }
    ]
  },
  "ownership_coverage": {
    "known_relationships": 3,
    "unknown_relationships": 227,
    "coverage_pct": 1.3
  },
  "other_subsidiaries": {
    "count": 227,
    "note": "Parent relationships unknown from public SEC filings"
  }
}

Ownership confidence levels:

root - Ultimate parent company
key_entity - Issuer or guarantor (relationship matters for credit analysis)
verified - Intermediate parent verified from indenture/credit agreement
unknown - From Exhibit 21 only (listed in other_subsidiaries, not in hierarchy tree)

Data Freshness Advantage

DebtStack extracts directly from the latest SEC EDGAR filings, providing 12-18 months fresher data than LLMs like ChatGPT or Gemini, which rely on stale training data.

Source	Data Period	NVDA EBITDA Example
DebtStack	Nov 2025 (FY2026 Q3)	$121B
Gemini/ChatGPT	Q1 2025 (knowledge cutoff)	$39B

This matters for fast-growing companies where financial metrics change significantly between LLM training updates.

Self-Hosting

Clone, setup, and run DebtStack locally

Clone and Setup

git clone https://github.com/debtstack-ai/debtstack-api.git
cd debtstack-api
pip install -r requirements.txt
cp .env.example .env
# Edit .env with your API keys

Database Setup

Neon Cloud (Recommended)

Create a free database at neon.tech

Copy the connection string to .env:

DATABASE_URL=postgresql+asyncpg://user:pass@host/db?sslmode=require

Run Migrations

alembic upgrade head

Run the API

uvicorn app.main:app --reload

Deployment

Railway (Recommended)

Connect GitHub: Link your repo at railway.app

Add Environment Variables:

DATABASE_URL=postgresql+asyncpg://...
ANTHROPIC_API_KEY=sk-ant-...
GEMINI_API_KEY=...
SEC_API_KEY=...

Deploy: Railway auto-deploys on push

See docs/DEPLOYMENT.md for detailed instructions.

Docker

docker-compose up -d

Environment Variables

Variable	Required	Description
`DATABASE_URL`	Yes	PostgreSQL connection string
`REDIS_URL`	Optional	Redis cache (Upstash)
`ANTHROPIC_API_KEY`	Yes	Claude API for escalation
`GEMINI_API_KEY`	Recommended	Gemini API for extraction
`SEC_API_KEY`	Recommended	SEC-API.io for filing retrieval
`FINNHUB_API_KEY`	Optional	Finnhub for bond pricing (FINRA TRACE)
`STRIPE_API_KEY`	Optional	Stripe for payment processing
`STRIPE_WEBHOOK_SECRET`	Optional	Stripe webhook verification
`SENTRY_DSN`	Optional	Sentry error tracking
`SLACK_WEBHOOK_URL`	Optional	Slack alerts for error spikes
`POSTHOG_API_KEY`	Optional	PostHog analytics (backend event tracking)
`POSTHOG_HOST`	Optional	PostHog host (default: `https://us.i.posthog.com`)

Extraction

Extract new companies using the iterative extraction system:

# Single company with QA (idempotent - safe to re-run)
python scripts/extract_iterative.py --ticker AAPL --cik 0000320193 --save-db

# Force re-run all steps (ignores skip conditions)
python scripts/extract_iterative.py --ticker AAPL --cik 0000320193 --save-db --force

# Batch extraction for all companies
python scripts/extract_iterative.py --all --save-db

# Resume batch from last processed company
python scripts/extract_iterative.py --all --save-db --resume

# Extract financials (single quarter)
python scripts/extract_financials.py --ticker CHTR --save-db

# Extract TTM financials (4 quarters - recommended for accurate leverage)
python scripts/extract_financials.py --ticker CHTR --ttm --save-db

# Recompute metrics after extracting financials
# Uses smart 10-K vs 10-Q logic:
#   - If latest filing is 10-K: uses annual figures directly (already TTM)
#   - If latest filing is 10-Q: sums trailing 4 quarters
python scripts/recompute_metrics.py --ticker CHTR

# Recompute metrics for all companies
python scripts/recompute_metrics.py

# Extract ownership hierarchy from Exhibit 21 HTML indentation
python scripts/extract_exhibit21_hierarchy.py --ticker CHTR --save-db

# Backfill treasury yields (free from Treasury.gov, needed for spread calculations)
python scripts/backfill_treasury_yields.py --from-year 2021 --to-year 2026

# Backfill historical bond pricing (requires Finnhub premium)
python scripts/backfill_pricing_history.py --all --days 1095 --with-spreads

The extraction pipeline is idempotent - safe to re-run on existing companies:

Skips steps where data already exists
Tracks extraction status (success/no_data/error) in database
Detects when new quarterly financials are available
Use --force to override skip logic and re-extract everything

Extraction steps (11 total):

Downloads 10-K, 10-Q, 8-K filings via SEC-API.io
Extracts entities and debt instruments with Gemini (~$0.008)
Runs 5 QA checks against source filings (~$0.006)
Applies targeted fixes if QA score < 85%
Escalates to Claude if still failing
Extracts document sections (indentures, credit agreements, debt footnotes)
Links debt instruments to their governing documents
Extracts TTM financials (4 quarters)
Parses Exhibit 21 for subsidiary hierarchy
Extracts guarantees and collateral (using linked documents)
Computes metrics and runs QC validation

Typical cost: $0.02-0.03 per company

Script Utilities

All scripts use shared utilities from scripts/script_utils.py for consistency:

from script_utils import get_db_session, print_header, run_async

async def main():
    print_header("SCRIPT NAME")
    async with get_db_session() as db:
        # Work with database session
        pass

if __name__ == "__main__":
    run_async(main())

Key utilities:

get_db_session() - Async database session with proper cleanup
run_async() - Runs async code with Windows event loop handling
print_header(), print_summary() - Consistent CLI output
create_base_parser() - Common CLI arguments (--ticker, --all, --limit)

Project Structure

credible/
├── app/
│   ├── api/
│   │   ├── primitives.py          # Primitives API (12 core endpoints)
│   │   ├── auth.py                # Auth API (signup, user info)
│   │   ├── pricing_api.py         # Pricing API (tiers, credits, usage)
│   │   ├── historical_pricing.py  # Historical bond pricing (Business)
│   │   ├── export.py              # Bulk export (Business)
│   │   ├── usage.py               # Usage analytics (Business)
│   │   └── routes.py              # Legacy REST endpoints
│   ├── core/
│   │   ├── config.py              # Configuration
│   │   ├── database.py            # Database connection
│   │   ├── cache.py               # Redis cache client
│   │   ├── auth.py                # API key generation, validation, tier config
│   │   ├── posthog.py             # PostHog analytics client
│   │   ├── monitoring.py          # Redis-based API metrics
│   │   ├── alerting.py            # Slack webhook alerts
│   │   └── scheduler.py           # APScheduler (pricing + alert jobs)
│   ├── models/schema.py           # SQLAlchemy models
│   └── services/
│       ├── # UTILITIES (stateless helpers)
│       ├── utils.py                 # Core: JSON parsing, name normalization
│       ├── extraction_utils.py      # SEC filing: HTML cleaning, content combining
│       ├── llm_utils.py             # LLM clients: Gemini, Claude, cost tracking
│       ├── yield_calculation.py     # Financial math: YTM, duration
│       │
│       ├── # SERVICES (orchestration)
│       ├── sec_client.py            # SEC filing clients (SecApiClient, SECEdgarClient)
│       ├── base_extractor.py        # Base class for LLM extraction services
│       ├── extraction.py            # ExtractionService + DB persistence
│       ├── iterative_extraction.py  # Main extraction with QA loop
│       ├── hierarchy_extraction.py  # Exhibit 21 parsing, ownership
│       ├── guarantee_extraction.py  # Guarantee relationships
│       ├── collateral_extraction.py # Collateral for secured debt
│       ├── covenant_extraction.py   # Structured covenant extraction
│       ├── qa_agent.py              # 5-check verification system
│       ├── financial_extraction.py  # Quarterly financials
│       ├── bond_pricing.py          # Pricing calculations
│       ├── pricing_history.py       # Historical pricing backfill
│       └── treasury_yields.py       # Treasury yield curves
├── scripts/                       # CLI tools
│   ├── script_utils.py            # Shared utilities (DB, parsers, progress)
│   ├── extract_iterative.py       # Complete extraction pipeline
│   ├── recompute_metrics.py       # Metrics recomputation
│   └── ...                        # See CLAUDE.md for full list
├── alembic/                       # Database migrations
├── docs/                          # Documentation
│   ├── DEPLOYMENT.md              # Deployment guide
│   ├── ACCOUNT_SETUP.md           # Vendor account setup
│   └── PRIMITIVES_API_SPEC.md     # Primitives API specification
└── results/                       # Extraction outputs

Documentation

docs/PRIMITIVES_API_SPEC.md - Primitives API specification with examples
docs/DEPLOYMENT.md - Full deployment guide for Railway
docs/ACCOUNT_SETUP.md - Step-by-step vendor account setup
sdk/README.md - Python SDK with LangChain and MCP integration guides
CLAUDE.md - AI assistant context for development

License

MIT

Name		Name	Last commit message	Last commit date
Latest commit History 205 Commits
alembic		alembic
app		app
demos		demos
docs		docs
medici		medici
scripts		scripts
sdk		sdk
tests		tests
.dockerignore		.dockerignore
.env.example		.env.example
.gitignore		.gitignore
CLAUDE.md		CLAUDE.md
Dockerfile		Dockerfile
README.md		README.md
WORKPLAN.md		WORKPLAN.md
alembic.ini		alembic.ini
docker-compose.yml		docker-compose.yml
pytest.ini		pytest.ini
railway.json		railway.json
railway.toml		railway.toml
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt
tier7_fixes.py		tier7_fixes.py

Folders and files

Latest commit

History

Repository files navigation

DebtStack.ai

Why DebtStack?

Quick Start

1. Get Your API Key

2. Install the SDK (or use curl)

3. Query

Current Database

Features

API Endpoints

Authentication

Primitives API (Optimized for AI Agents)

Advanced Primitives

Traverse Entities - POST /v1/entities/traverse

Search Documents - GET /v1/documents/search

Agent Workflow: Discovery → Deep Dive

System

Python SDK

MCP + LangChain

Pricing

Data Quality

Estimated Data Transparency

Leverage Ratio Data Quality

Ownership Relationship Transparency

Data Freshness Advantage

Self-Hosting

Clone and Setup

Database Setup

Run Migrations

Run the API

Deployment

Railway (Recommended)

Docker

Environment Variables

Extraction

Script Utilities

Project Structure

Documentation

License

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Traverse Entities - `POST /v1/entities/traverse`

Search Documents - `GET /v1/documents/search`

Packages