Integration Pathways
Twelve persona workflows across HTTP API, batch data pipelines, CLI, Python SDK, and MCP — all calling the same deterministic engine.
Persona workflows · Choose your surface · Workflows · HTTP API reference · Why deterministic scoring
Persona Workflows
Run disclosure-alpha-api locally. Same deterministic scores as CLI — REST for dashboards, services, and research stacks.
01 / METRICS
Quant / researcher
GET /v1/company/{ticker}/disclosure-metrics
Tone ratios, boilerplate proxies, boolean risk flags, and section diffs without score aggregation. Use compare=prior|none and optional sections= filters.
RAW
METRICS + FLAGS + DIFFS
02 / RISK
Risk analyst
GET /v1/company/{ticker}/disclosure-matrix
Headline overall_disclosure_risk_score (0–100), nine headline components, score_coverage_ratio, and missing_components. tier=standard or analyst adds metrics and provenance.
0–100
RISK SCORE RANGE
03 / DASHBOARD
Dashboard / product engineer
GET …/disclosure-matrix?tier=lite
Minimal JSON with overall_disclosure_risk_score only — ideal for dashboard tiles, watchlists, and internal tools that need one number per ticker.
lite
HEADLINE ONLY
04 / COMPLIANCE
Compliance / audit analyst
GET …/disclosure-flags · GET …/disclosure-changes
Boolean risk events (material weakness, restatement, going concern) or section-level year-over-year diff detail — without recomputing the full disclosure matrix.
2
FLAGS + CHANGES
Choose your surface
Pick an input type. We'll point you to the right entry point.
Building with MCP?
Quick Start Recipes
pip install "disclosure-alpha" export SEC_USER_AGENT="YourName your@email.com" disclosure-alpha score --ticker AAPL --fiscal-year 2025 --form 10-K pip install "disclosure-alpha" disclosure-alpha score --html tests/fixtures/sample_10k.html --form 10-K jq '{overall: .scores.overall_disclosure_risk_score, coverage: .scores.score_coverage_ratio}' Save current and prior 10-K HTML locally disclosure-alpha score --html current.html --form 10-K --prior-html prior.html Inspect disclosure_change_score in JSON output pip install "disclosure-alpha[api]" disclosure-alpha-api (listens on :8000) GET /v1/company/AAPL/disclosure-matrix?fiscal_year=2025&form_type=10-K&tier=lite disclosure-alpha-api POST /v1/panel/disclosure-matrix with tickers array (max 25) Sort results by overall_disclosure_risk_score from disclosure_alpha import score_filing_ticker result = score_filing_ticker("AAPL", 2025, form_type="10-K") for name, value in result.scores.components.dict.items(): print(name, value) disclosure-alpha-api GET /v1/company/{ticker}/disclosure-flags?fiscal_year=2025&form_type=10-K Review boolean risk events without full matrix disclosure-alpha-api GET /v1/company/AAPL/sections?fiscal_year=2025§ions=item_1a_risk_factors Feed section text into your warehouse or NLP stack pip install "disclosure-alpha[mcp]" Add disclosure-alpha-mcp-analyst to your MCP server config Set SEC_USER_AGENT in env; call score_company_filing_tool Reference
| Surface | Best for | Entry point | Install |
|---|---|---|---|
| CLI | Terminal workflows, local HTML | disclosure-alpha | (base) |
| Python SDK | Custom pipelines, notebooks | import disclosure_alpha | (base) |
| HTTP API | Services, dashboards, screeners | disclosure-alpha-api | [api] |
| OpenBB Workspace | Analysts in OpenBB Workspace | disclosure-alpha-api | [api,mcp] |
| MCP Analyst | Agent ticker workflows | disclosure-alpha-mcp-analyst | [mcp] |
| MCP Builder | Raw HTML agent workflows | disclosure-alpha-mcp-builder | [mcp] |
Install from PyPI and score your first filing in minutes.