vibe-trading

Professional finance research toolkit — backtesting (10 engines + benchmark comparison panel), factor analysis, Alpha Zoo (462 pre-built alphas across qlib158/alpha101/gtja191/academic/fundamental), options pricing, 90 finance skills, 30 multi-agent swarm teams, Trade Journal analyzer, and Shadow Account (extract → backtest → render) across 28 market-data sources (tushare, yfinance, okx, binance, akshare, baostock, tencent, mootdx, ccxt, futu, mt5, tickerall, local, eastmoney, sina, stooq, yahoo, pykrx, india_broker, qveris, longbridge, nobitex, wallex, plus optional-key finnhub/alphavantage/tiingo/fmp/gildata).

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Failed to scan

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./agent/SKILL.md

Quality

Content

60%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is information-dense and largely actionable with clear step sequences for the flagship workflows, but it is a monolith: hundreds of lines of tool-inventory, market-matrix, and connector-configuration detail that belong in one-level-deep reference files are inlined, and several regulatory-detail paragraphs are run-on and over-long. Splitting into references and trimming per-market micro-detail would address both weaknesses.

Suggestions

Move the 76-tool table, the market-source matrix, and the per-market engine/regulatory details into separate reference files (e.g. references/tools.md, references/markets.md, references/connectors.md) linked from short overview sections in SKILL.md.

Show a minimal working config.json example (and the signal_engine.py shape) for the core backtest workflow so the 'write_file() to create config.json' step is fully executable.

Tighten the Vietnam/Korea/India paragraphs into short bullets (suffix convention, cost model, key constraints) and move the regulatory micro-detail to the per-market reference file.

Dimension	Reasoning	Score
Conciseness	Mostly information-dense project-specific detail Claude would not already know (engines, connectors, cost stacks), but several sections bury micro-detail in run-on sentences — e.g. the Vietnam and Korea paragraphs pack settlement cycles, tick grids, and tax rates into single 100+ word sentences — and could be tightened considerably. It is not anchor-2 padding of known concepts, but it is noticeably over-long.	3 / 5
Actionability	Mostly executable guidance: copy-paste 'pip install vibe-trading-ai', complete MCP config JSON blocks, concrete CLI commands ('vibe-trading alpha bench --zoo gtja191 --universe csi300 --period 2018-2025', connector profile sequence), and parameterized example prompts. Minor gap: the core backtest workflow says 'write_file() to create config.json' without ever showing a config.json example or its schema.	4 / 5
Workflow Clarity	The flagship Shadow Account loop is a clearly numbered 5-step sequence, the backtest example workflow is a 4-step sequence, and the connector flow includes an explicit 'connector check' validation checkpoint plus a failure-handling table. Not anchor 5 because the main backtest and swarm workflows have no explicit validation or error-recovery steps — checkpoints are implicit or absent outside the connector path.	4 / 5
Progressive Disclosure	A ~390-line monolithic document: the 76-row tool inventory, the market-by-market source matrix, the per-market engine/regulatory details, and the external-MCP/connector configuration docs all clearly belong in separate reference files, yet no references exist — the body links to nothing. The bundled scripts/ files are development-only and never referenced. Structure within the file is decent, but content that clearly belongs in separate files is inlined, matching anchor 2.	2 / 5
	Total	13 / 20 Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A highly specific, distinctive description that comprehensively inventories the toolkit's capabilities, but it is bloated by a 24-source enumeration and completely lacks a 'Use when...' trigger clause. Adding explicit usage triggers and trimming the source list would lift completeness and specificity substantially.

Suggestions

Add an explicit trigger clause, e.g. 'Use when backtesting trading strategies, running factor/alpha analysis, pricing options, or analyzing a trade journal across A-share/HK/US/crypto markets.'

Trim the 24-name data-source enumeration to the handful users actually name (yfinance, tushare, ccxt) or move it to the body — it inflates the description without adding trigger value.

Include one or two common user synonyms ('quant research', 'trading strategy') to widen natural trigger coverage.

Dimension	Reasoning	Score
Specificity	The description lists many concrete capabilities ('backtesting (10 engines + benchmark comparison panel)', 'factor analysis', 'options pricing', 'Shadow Account (extract → backtest → render)') with comprehensive toolkit coverage, but it reads as an overstuffed feature inventory — the 24-name data-source enumeration pads rather than clarifies — rather than the crisp verb-led action list of the anchor-5 example.	4 / 5
Completeness	The 'what' is clear and detailed, but there is no 'Use when...' clause or equivalent explicit trigger guidance anywhere — the when-to-use question is entirely unanswered, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Good natural domain keywords ('backtesting', 'factor analysis', 'options pricing', 'Trade Journal analyzer') plus library names users actually say (yfinance, tushare, ccxt), but common variations like 'quant', 'strategy', or 'portfolio' are missing, so it falls short of the comprehensive synonym coverage of anchor 5.	4 / 5
Distinctiveness Conflict Risk	It occupies a clear professional finance-research niche (backtest engines, alpha zoos, broker-journal analysis, named market-data sources) with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	16 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/SKILL.md
Commit: 251b094

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.