agentsclimarketplace

Finmind

Skill chenwei791129/agent-skills/skills/finmind

Query Taiwan stock financial reports via FinMind API. Use when: (1) analyzing Taiwan company financials (income statement, balance sheet, cash flow), (2) checking monthly revenue trends, (3) looking up dividend history or PER/PBR/yield, (4) comparing company financial metrics, (5) any question about Taiwan stock fundamentals. Trigger on stock codes like 2330, 2317, or company names like TSMC, Hon Hai, MediaTek.From its SKILL.md

Install
npx -y skills add chenwei791129/agent-skills --skill finmind

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

6.9 KB, ~1.9k tokens by cl100k_base, as published. Nobody here has run it

FinMind Taiwan Financial Report Query

Query Taiwan listed/OTC company financial data via FinMind REST API. Covers income statements, balance sheets, cash flows, monthly revenue, dividends, PER/PBR, and stock prices.

Prerequisites

Optional: Set FINMIND_TOKEN for higher rate limits (600 req/hr vs 300 req/hr without token).

Register at https://finmindtrade.com/ to get a free token, then:

export FINMIND_TOKEN="your-token-here"

Available Commands

All commands are run via:

uv run ~/.claude/skills/finmind/scripts/finmind_query.py <command> [options]

Financial Data Commands

CommandDescriptionKey Fields
incomeComprehensive Income StatementEPS, revenue, gross profit, operating income, net income
balance-sheetBalance Sheettotal assets, liabilities, equity, cash
cash-flowCash Flow Statementoperating/investing/financing cash flows
revenueMonthly Revenuemonthly revenue, YoY growth, MoM growth
dividendDividend Policycash dividend, stock dividend
perPER / PBR / Dividend Yieldprice-to-earnings, price-to-book, yield
priceDaily Stock Priceopen, high, low, close, volume

Utility Commands

CommandDescription
datasetsList all available FinMind datasets
translate --dataset <name>Show Chinese-English field name mapping

Date Range Guidelines

IMPORTANT: Choose --start dynamically based on today's date and the analysis goal. Do NOT hardcode dates from examples — calculate them relative to the current date.

Analysis GoalRecommended --startReasoning
Recent quarterly financials2 years ago (e.g. today minus 2Y)Captures ~8 quarters for trend analysis
Monthly revenue trend1 year ago12 months shows seasonal patterns
Dividend history5 years agoEnough to assess dividend growth stability
PER/PBR valuation3 months ago with --limit 30Recent trading days for current valuation
Stock price1-3 months ago with --limit 20Recent price action
Full fundamental deep-dive3 years agoComprehensive multi-year trend

Example: if today is 2026-04-06, use --start 2024-04-06 for 2-year income analysis, not a fixed date.

Usage Examples

Query income statement (recent 2 years)

uv run ~/.claude/skills/finmind/scripts/finmind_query.py income -s 2330 --start <2-years-ago>

Query balance sheet with row limit

uv run ~/.claude/skills/finmind/scripts/finmind_query.py balance-sheet -s 2330 --start <2-years-ago> --limit 20

Query monthly revenue (recent 1 year)

uv run ~/.claude/skills/finmind/scripts/finmind_query.py revenue -s 2317 --start <1-year-ago>

Query PER/PBR/yield (recent)

uv run ~/.claude/skills/finmind/scripts/finmind_query.py per -s 2330 --start <3-months-ago> --limit 30

Output as JSON (for programmatic processing)

uv run ~/.claude/skills/finmind/scripts/finmind_query.py income -s 2330 --start <2-years-ago> --format json

Translate field names

uv run ~/.claude/skills/finmind/scripts/finmind_query.py translate --dataset TaiwanStockFinancialStatements

Common Arguments

ArgumentDescription
--stock, -sStock ID (e.g. 2330) — required for data commands
--startStart date YYYY-MM-DD — required for data commands
--endEnd date YYYY-MM-DD (default: today)
--format, -fOutput format: table (default) or json
--tokenAPI token (overrides FINMIND_TOKEN env var)
--limit, -lLimit number of rows returned

Common Stock Codes

CodeCompanyIndustry
2330TSMC (台積電)Semiconductor
2317Hon Hai / Foxconn (鴻海)Electronics Manufacturing
2454MediaTek (聯發科)IC Design
2308Delta Electronics (台達電)Power Supplies
2881Fubon FHC (富邦金)Financial Holding
2882Cathay FHC (國泰金)Financial Holding
2891CTBC FHC (中信金)Financial Holding
2886Mega FHC (兆豐金)Financial Holding
2412Chunghwa Telecom (中華電)Telecom
1301Formosa Plastics (台塑)Petrochemical
2303UMC (聯電)Semiconductor
3711ASE Technology (日月光投控)Semiconductor Packaging
2002China Steel (中鋼)Steel
1216Uni-President (統一)Food
2382Quanta Computer (廣達)Server / Cloud

Analysis Patterns

Use these query combinations for common analysis scenarios:

Full Fundamental Analysis

# Run these 3 commands in parallel (use --start <3-years-ago>):
uv run ~/.claude/skills/finmind/scripts/finmind_query.py income -s 2330 --start <3-years-ago>
uv run ~/.claude/skills/finmind/scripts/finmind_query.py balance-sheet -s 2330 --start <3-years-ago>
uv run ~/.claude/skills/finmind/scripts/finmind_query.py cash-flow -s 2330 --start <3-years-ago>

Revenue Momentum Tracking

uv run ~/.claude/skills/finmind/scripts/finmind_query.py revenue -s 2330 --start <1-year-ago>

Dividend Investment Assessment

uv run ~/.claude/skills/finmind/scripts/finmind_query.py dividend -s 2330 --start <5-years-ago>
uv run ~/.claude/skills/finmind/scripts/finmind_query.py per -s 2330 --start <3-months-ago> --limit 30

Quick Valuation Check

uv run ~/.claude/skills/finmind/scripts/finmind_query.py per -s 2330 --start <3-months-ago> --limit 10
uv run ~/.claude/skills/finmind/scripts/finmind_query.py price -s 2330 --start <3-months-ago> --limit 10

Multi-Company Comparison

# Compare revenue of TSMC vs UMC:
uv run ~/.claude/skills/finmind/scripts/finmind_query.py revenue -s 2330 --start <1-year-ago>
uv run ~/.claude/skills/finmind/scripts/finmind_query.py revenue -s 2303 --start <1-year-ago>

Rate Limits

TierLimitHow to Get
No token300 requests/hourDefault
With token600 requests/hourRegister at https://finmindtrade.com/

When rate-limited (HTTP 402), the script will suggest setting a token.

Data Coverage

  • Income Statement: from 1990-03-01
  • Balance Sheet: from 2011-12-01
  • Cash Flow Statement: from 2008-06-01
  • Monthly Revenue: from 2002-02-01
  • Dividend Policy: from 2005-05-01
  • PER/PBR/Yield: varies
  • Stock Price: varies

Data is updated daily by FinMind.

What ships with it: 1 file

7.3 KB alongside SKILL.md, 1 of them executable

scripts/

Gives 0 of the 12 instructions most finance skills give in ~1.9k tokens

Counted across 469 of the 469 authors here whose files we hold, read 2026-08-07

  • Extract date vendor amount and descriptionin 15 of 469, across 3 files
  • Scan folder for invoice filesin 14 of 469, across 2 files
  • Rename files to standard formatin 14 of 469, across 2 files
  • Show organization plan before movingin 14 of 469, across 2 files
  • Generate summary CSVin 14 of 469, across 2 files
  • Organize files by categoryin 13 of 469, across 1 file
  • Preserve original filesin 13 of 469, across 1 file
  • Flag files missing critical infoin 13 of 469, across 1 file
  • Produce the requested output filein 9 of 469, across 4 files
  • Build best, base, and worst case scenariosin 9 of 469, across 5 files
  • Implement backoff if rate limit errors occurin 8 of 469, across 3 files
  • Determine the weighted average cost of capitalin 8 of 469, across 4 files

Said here and by no other author read

  • Calculate start dates dynamically relative to today
  • Do not hardcode dates from examples
  • Run independent queries in parallel
  • Provide a stock ID for data commands
  • Provide a start date for data commands
  • Set token environment variable for higher rate limits

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.