TW Market Data LogoTW Market Data

Facts & statistics

Market facts

Long-run statistics for Taiwan equities, each with its sample period and as_of

Seasonality

Monthly return distributions and hit rates, with sample sizes

Institutional flow

Seasonality in the breadth of institutional net buying

Limit events

How often limit moves happen, and how concentrated they are

Delisting

Delisting counts and survival spans — the basis for avoiding survivorship bias

Rule changes

A timeline of trading-rule changes — the premise for reading historical data

Data & exploration

Dataset catalogue

Every dataset, its coverage, and how often it updates

Playground

Call the API from the browser, without a key

Market today

Today's market at a glance

Stock analysis

The entry point for looking at one instrument

Market heat map

The whole market in one picture — area is market cap, colour is revenue growth

Market calendar

Statutory disclosure deadlines — the day a figure may legally first be known

Platform capabilities

Product overview

What TWMD provides, and who it is built for

Verifiable proof

Signed checkpoints and per-row inclusion proofs

Data quality

Reconciliation, gap handling, and quality status

Methodology

How the figures are computed, and on what basis

Auditable execution

Tie a trading decision back to the data it saw

Connect your broker

Bring TWMD into an existing order and research workflow

Developers

Documentation

API reference, dataset pages, and integration guides

Quick start

Authentication and your first request

Integrate by role

Separate paths for quant research, data engineering, and app development

Connect over MCP

Point an agent straight at TWMD

MCP registry

The published MCP tool list and its signed manifest

Webhooks

Have your system told when data updates

Learn

Blog

Long-form writing on data, method, and market structure

Answers

Specific answers to specific questions, with sources

Topics

Industry chains and thematic relationships

Help centre

Account, billing, and usage questions

Glossary

Definitions for Taiwan-market and data terms

Compare & status

Why TWMD

A point-by-point comparison with FinMind and TEJ

Migrate from FinMind

Field mapping and migration steps

Migrate from FinLab

Field mapping and migration steps

Migrate from TEJ

Field mapping and migration steps

Status

Service availability and incident history

Security & trust

Trust centre

What we claim, and the limits on each claim

Security

Architecture, access control, and incident handling

Security facts

The items you can verify from outside

Security evidence

SBOM, ASVS mapping and threat model — including the three controls we do not meet.

Self-assessment

Item-by-item answers for a procurement questionnaire

Compliance & standards

Compliance mapping

Evidence primitives mapped onto FSB, IOSCO, and SR 26-2

Standards & interop

Term-by-term mapping onto published standards, and where it does not map

Provenance & C2PA

A machine-readable origin graph, fetchable without a key

Licensing

How the data may be used and redistributed

Adoption

Evaluate

Seven checks you can run yourself, without an account

Talk to sales

Enterprise plans, quotas, and contract detail

Pricing
中文Sign inSign up
TW Market Data

Taiwan market-data infrastructure, built for AI agents and quantitative workflows.

Status unknown
© 2026 TW Market Data

TW Market Data (TWMD) provides historical data and statistics, not investment advice; investment decisions and their risks are your own.

  • Privacy Policy·
  • Terms of Service·
  • Cookie Policy·
  • Acceptable Use Policy·
  • Data Sources & Licensing·
  • Legal (all documents)
中文

AI Agent

  • MCP Server
  • Skills
  • Tool manifest
  • Agent workflow examples
  • Agent benchmark
  • llms.txt
  • OpenAPI spec

Security

  • Security overview
  • Verifiable data
  • Trust Center
  • Standards & Interop
  • Regulatory mapping

Product

  • Datasets
  • Topics
  • Market facts
  • Documentation
  • Integration runbooks
  • Playground (no signup)
  • Free tier
  • Solutions

Company

  • About TWMD
  • Blog
  • Help centre
  • Pricing

Documentation

DASHBOARD

DashboardPricing

FOR AI AGENTS

MCP ServerSkillsllms.txtTool manifestOpenAPI SpecAgent workflow examples

OVERVIEW

OverviewQuick startAuthenticationSource policyData gradesData lineageMarket coverage

DATA APIS

GUIDES

How to get the 3 financial statementsHow to read institutional flowsHow to check market statusHow to wire a strategy / AI agent

SDKS

Release statusPython SDKJavaScript / TypeScript SDK

Building an AI agent? Start with /llms.txt for the full site index.

Docs

Agent workflow examples

A simple research workflow, showing how several dataset queries combine into structured research context.

Point-in-time: keeping an agent's backtest from seeing the future

Every row of disclosure data carries a knowledge date, set conservatively to the statutory filing deadline — so a backtest filtered on it does not see numbers the market could not see at the time.

Every disclosure row carries a knowledge date. Monthly revenue, the comprehensive income statement, the balance sheet and the cash flow statement all carry knowledge_date, alongside kd_source explaining where that date came from and kd_imputed marking whether it was observed or derived. Align on that column rather than on the period end date, and a number enters your backtest when it entered the market, not when the quarter closed.

It is the statutory deadline, and it is deliberately late. Measured 2026-08-19: on every disclosure dataset checked, kd_source was statutory_deadline and kd_imputed was true. That date is the legal filing deadline, not the moment of actual announcement, which we do not yet capture. Companies file on or before the deadline, so the date runs late rather than early: a company that announced ahead of schedule looks slower than it was. That costs some fidelity, but it cannot leak future information, and that is the safer direction for the bias to run.

Daily prices do not have this column. twse_daily_price carries no knowledge_date field. For a closing price the trading day is the publication day, so the gap is small — but it is still a real gap, and an agent that assumes the column exists everywhere gets undefined rather than a date. The point-in-time properties of each dataset are listed on the coverage page.

as-of truncation over REST. Measured 2026-08-22: the REST dataset endpoints apply it. Passing as_of moves the latest row back to the cut-off — 2330 daily prices return 2026-08-21 with no parameter and 2020-06-30 with as_of=2020-06-30. The response reports as_of_applied, which field was actually filtered, and how many rows were cut, so you can confirm it worked rather than take our word for it. This is truncation on **knowledge time**, not on date: with as_of=2021-03-31 monthly revenue returns February, because February's revenue is not filed until March. On filing-based datasets that knowledge time is the **statutory filing deadline** rather than the actual announcement date, so data never appears earlier than it genuinely became public: conservative, safe for backtesting, and not precise. Daily prices have no filing deadline and are truncated on the trading day.

What this is

This example is not an AI model. It is a financial data retrieval workflow.

The point is to show how an agent uses TW Market Data as its data tool layer.

Example files

  • examples/agents/simple_research_agent.ts
  • examples/agents/simple_research_agent.py

The flow

  • Read TWMD_API_KEY
  • Query issuer_profile
  • Query twse_daily_price
  • Query monthly_revenue
  • Emit structured research context

Traceability

Every step returns a requestId and credits metadata, which is what makes the run observable and a support question answerable.

Running it locally

TWMD_API_KEY=sk_live_xxx node examples/agents/simple_research_agent.tsTWMD_API_KEY=sk_live_xxx python3 examples/agents/simple_research_agent.py

On this page

  • Point-in-time: keeping an agent's backtest from seeing the future
  • What this is
  • Example files
  • The flow
  • Traceability
  • Running it locally