Skip to content

Guide · Agent skill

How to use the QVeris Paid Data Marketplace agent skill

Updated · written with AI from the project's own documentation and checked against our catalogue data

The qveris skill teaches an AI agent to use QVeris, a paid marketplace of data providers covering stocks, options, macro series, crypto, China/HK data, news, filings and alternative data. It is part of the HKUDS/Vibe-Trading project and is released under the MIT licence. This guide explains what the skill does, how to install it, how to ask for it in plain language, and what to watch for because paid calls spend real credits.

QVeris Paid Data MarketplaceHKUDS/Vibe-Trading · MIT

Quick start

  1. 1Get an account and API key. Sign up through the QVeris link on the skill page, or use the invite code if asked. Copy your API key.
  2. 2Install Vibe-Trading and the skill. Make sure Vibe-Trading is installed. Download the skill ZIP and unzip it into ~/.claude/skills/ or the project's .claude/skills/ folder, or upload it under Settings → Capabilities → Skills.
  3. 3Set the API key. Set QVERIS_API_KEY, and QVERIS_BASE_URL if needed, or enter the key under Settings → QVeris.
  4. 4Enable paid mode. Run vibe-trading data mode paid so the agent can use QVeris discovery, inspection and execution.
  5. 5Ask for the data by name. Make a concrete request that names QVeris, such as an options chain with Greeks for a specific ticker.
  6. 6Check the cost. Review the success rate and expected cost the agent reports, then confirm the call. Afterwards, check cost and remaining credits.

What the qveris skill makes the agent do

By default, an agent using Vibe-Trading gets price data (OHLCV) from free public sources. The qveris skill adds a paid route for cases where those free sources cannot cover the dataset, do not have enough depth, or lack a provider of the quality you need. Examples the source names include options Greeks, analyst and calendar feeds, broad provider comparison, and paid China/HK or global coverage.

The skill sets a strict order of operations. The agent searches for candidate capabilities first, inspects the one it wants, and only then runs a billable call. Throughout, it is meant to surface cost and success rate so you can see what is being spent.

  • qveris_search: finds candidate capabilities. Search is free.
  • qveris_inspect: returns the full parameter schema and sample parameters for a chosen tool_id. Inspect is free.
  • qveris_execute: runs the billable call with the exact parameters, and reports cost and remaining_credits.

Requirements before you install

The skill needs a working Vibe-Trading installation, a QVeris account with an API key, and paid credits on that account for billable calls. You can sign up through the link on the skill page, or use the invite code Vyjjo5G_1cAHJA if signup asks for one. Without an API key the QVeris tools are treated as unavailable.

  • Vibe-Trading installed
  • QVeris account and API key, set as QVERIS_API_KEY
  • Optional: QVERIS_BASE_URL, if you need a non-default endpoint
  • Paid credits on the account for billable calls
QVERIS_API_KEY=your_key_here

Install the qveris skill from the ZIP

Download the ZIP for the skill from our page. For a personal install, unzip it into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder instead. The unzipped folder should contain SKILL.md, plus references/coverage.md and references/rest-api.md.

If you use the Claude apps rather than a folder on disk, you can upload the ZIP instead. Go to Settings, then Capabilities, then Skills, and add it there.

  • Personal install: ~/.claude/skills/
  • Project install: the project's .claude/skills/ folder
  • Claude apps: Settings → Capabilities → Skills

Turn on the paid route and set a budget

QVeris is off in free mode. To allow paid discovery, inspection and execution, set the mode to paid. You can do this in Settings under QVeris, or with the command below. In paid mode, calls are limited by budget_credits_per_session, so set that limit to what you are willing to spend in one session.

You can supply the key through the environment variables or through Settings. The source lists both routes.

  • Settings → QVeris
  • Environment variables: QVERIS_API_KEY and QVERIS_BASE_URL
  • Only flat per-call quotes can be reserved against the budget. Quotes priced per result, row or value are refused as quote_not_bounded.
vibe-trading data mode paid

Trigger the skill with natural requests

You do not need to type tool names. Ask for the data you need and name QVeris if you want to be sure the paid route is used. The skill is also triggered when the free sources cannot cover the request. Concrete queries work best, because the agent passes them to search. The source gives these examples.

  • Use QVeris to get the US listed options chain with implied volatility and Greeks for AAPL.
  • Get Hong Kong daily OHLCV and financial statements through QVeris.
  • Pull FRED CPI and the Treasury yield curve from QVeris.
  • Find China A share northbound fund flow daily data using QVeris.

Read cost, success rate and balance before you pay

Providers in the catalogue vary widely in quality. The source records some tools with near-zero success and others with 0%, so a paid result can be empty or poor. When you are choosing between paid options, the agent should show each candidate's success rate and expected cost. Failed calls, empty results and parameter errors are not charged under the QVeris contract.

Every successful execute result should include cost and remaining_credits. To check your balance and recent usage, use the web settings or the CLI commands below.

  • Web: Settings → QVeris
  • CLI status: vibe-trading data status
  • CLI usage: vibe-trading data usage
  • For reconciliation, compare execution_id, tool_id, charge_outcome, cost and remaining_credits.
vibe-trading data status

Handle large responses and truncation

qveris_execute accepts max_response_size. The default is 20480, and -1 asks QVeris not to truncate. If a response is too large, you receive a message, truncated_content and full_content_file_url. If you need the full JSON, the agent fetches the signed URL. If a summary is enough, it should say the response was truncated and work only from truncated_content.

  • Default max_response_size: 20480
  • -1 means do not ask QVeris to truncate

What people use it for

  • Pulling an options chain with implied volatility and Greeks for a US listed stock.
  • Getting analyst or calendar feeds that free sources do not provide.
  • Retrieving Hong Kong or China A share data for a backtest that free sources do not cover.
  • Comparing several paid providers for the same dataset by success rate and cost before spending credits.
  • Pulling FRED macro series such as CPI or the Treasury yield curve.

Frequently asked questions

Is the qveris skill free to use?

The skill itself is MIT licensed. Searching and inspecting capabilities are free, but billable execute calls spend credits from your QVeris account. Routine OHLCV price data stays on Vibe-Trading's free sources by default.

What do I need to run the qveris skill?

You need a working Vibe-Trading installation, a QVeris account with an API key set as QVERIS_API_KEY, and paid credits for billable calls. QVERIS_BASE_URL is optional.

How do I install the qveris skill?

Download the ZIP from the skill page and unzip it into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder. In the Claude apps you can instead upload it under Settings → Capabilities → Skills.

Why does the agent say the QVeris tools are unavailable?

The source says the tools are unavailable if QVeris is disabled, the mode is free, or no API key is configured. Check your key, then run vibe-trading data mode paid.

Will I be charged if a paid call returns nothing?

Failed calls, empty results and parameter errors are not charged under the QVeris contract. Successful execute results do show cost and remaining_credits so you can track spending.

Should I use qveris instead of free data sources?

Not for routine daily OHLCV work. The skill says to keep free sources as the default and use QVeris when they lack coverage, depth or provider quality.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

The qveris skill suits analysts and quant researchers who already have a QVeris account and budget, and who need options Greeks, analyst feeds or paid China/HK or global coverage. It is a poor fit for routine price data, for users without credits, or for anyone expecting every paid provider to return good data. Set a session budget and check the success rate before you run paid calls.

Download the skill