Skip to content
Get packs

Guide · Agent skill

How to use the Seasonal Calendar Trading Signals agent skill

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

The seasonal skill teaches an AI agent to build a simple rules-based trading strategy from calendar effects, such as certain months or weekdays historically performing better or worse. It is aimed at quant researchers, developers and students who want to test whether these time-based patterns hold in a given market. After this guide you will know what the skill does, how to install it, how to ask for it in plain language, and what to check before trusting its output.

Seasonal Calendar Trading SignalsHKUDS/Vibe-Trading · MIT

Quick start

  1. 1Download and unzip. Download the skill ZIP from this page and unzip it to get SKILL.md and example_signal_engine.py.
  2. 2Place the skill folder. Move the folder into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. Alternatively, upload it under Settings, Capabilities, Skills.
  3. 3Install dependencies. Run pip install pandas numpy so the Python example engine can run.
  4. 4Prepare your OHLCV data. Have a price file with open, high, low, close and volume data indexed by date.
  5. 5Ask for signals. Describe the bullish and bearish months or weekdays you want, and ask the agent to label each period as 1, -1 or 0.
  6. 6Backtest the result. Check the sample size and test the signals independently before relying on them.

What does the seasonal skill make the agent do?

An agent skill is a folder of instructions that an AI agent loads when it needs a specific capability. This one covers seasonal and calendar-effect strategies. It uses time-based regularities in financial markets, such as month-of-year effects and day-of-week effects, to generate trading signals.

Each period is labelled as 1 (long, bullish window), -1 (short, bearish window) or 0 (stand aside). The skill uses pandas and numpy and works with any OHLCV price data you supply. It describes the logic and includes a Python example engine.

  • Month effect (default): bullish months give long signals, bearish months give short or flat signals, all other months stay flat.
  • Day-of-week overlay (optional): Monday and Friday effects, plus start-of-month and end-of-month effects.
  • Combined mode: a position opens only when the month signal and the weekday signal both confirm.

How do you install the seasonal skill?

Download the ZIP for this skill from this page and unzip it. For a personal setup, place the skill folder in ~/.claude/skills/. For a single project, place it in the project's .claude/skills/ folder.

You can also upload the skill in the Claude apps. Go to Settings, then Capabilities, then Skills, and upload the skill there.

  • Files included: SKILL.md and example_signal_engine.py.
  • Install the Python dependencies with: pip install pandas numpy
pip install pandas numpy

How do you trigger the skill with natural requests?

You do not need special syntax. Describe the calendar pattern you want to test and the market data you have, and the agent will use the skill when the request matches its description. Say which months or weekdays you consider bullish or bearish, or ask for the defaults.

The skill sets the default bullish months to January, February, March, November and December, and the default bearish months to May through September. Weekday effects are off by default and must be turned on.

  • "Use the seasonal skill to label each month of my OHLCV file as long, short or flat."
  • "Set bullish months to 1, 2 and 3 and bearish months to 5 through 10, then generate signals."
  • "Add a Friday bullish and Monday bearish weekday overlay and show me the signals."

What parameters can you set?

The skill exposes a small set of parameters. You can change which months and weekdays count as bullish or bearish, and whether weekday effects are used at all.

  • bullish_months: default [1, 2, 3, 11, 12]
  • bearish_months: default [5, 6, 7, 8, 9]
  • use_weekday: default False
  • bullish_weekdays: default [4] (Friday)
  • bearish_weekdays: default [0] (Monday)

What are the common pitfalls?

Most errors come from indexing. Pandas numbers months from 1 (January) and weekdays from 0 (Monday) to 4 (Friday). Neutral months, which appear in neither the bullish nor the bearish list, should output 0 and must not be skipped.

Calendar effects are statistical regularities, not deterministic signals. Check the sample size in any backtest before drawing conclusions.

  • pd.DatetimeIndex.month starts at 1 for January.
  • pd.DatetimeIndex.weekday starts at 0 for Monday.
  • Neutral periods must output 0, not be dropped.

What are the safety notes?

This skill runs scripts, through its Python example engine. It does not use the network. The skill provides no risk management, no position sizing and no transaction cost modelling, so none of these are handled for you.

Do not use the output for live trading without independent backtesting and your own risk controls.

  • Runs scripts: yes (example_signal_engine.py).
  • Uses network: no.
  • Requires your own OHLCV price data.

What people use it for

  • Testing whether the sell-in-May pattern held in a given index's price history.
  • Checking whether a spring rally window (January to March) shows up in China A-share data.
  • Prototyping a simple rules-based strategy from month and weekday labels.
  • Teaching students how calendar anomalies are coded and tested with pandas.

Frequently asked questions

How do I install the seasonal agent skill?

Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. You can also upload it in the Claude apps under Settings, then Capabilities, then Skills.

What does the seasonal skill need to run?

It needs Python, pandas and numpy, plus OHLCV price data that you supply. Install the libraries with pip install pandas numpy.

Is the seasonal agent skill free?

The skill is published under the MIT licence, as stated on this page. Check the licence file in the source repository for the full terms.

Does this skill predict the market?

No. It labels periods based on historical calendar patterns. These are statistical regularities that can fail or disappear, so they need backtesting.

What is the difference between month effect and day-of-week effect here?

The month effect is on by default and labels whole months as bullish, bearish or flat. The day-of-week effect is optional and turns on with use_weekday. In combined mode, a position opens only when both signals confirm.

Why does my signal output show zeros?

Zeros are expected. Months that are neither bullish nor bearish output 0, meaning stand aside. Make sure these rows are kept rather than dropped, and check that months use the 1 to 12 numbering.

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

Use this skill if you are a quant researcher, developer or student who wants to test calendar effects on OHLCV data and see how a simple rules-based signal is coded. Pick something else if you want reliable profits or a live trading system, because the skill has no risk management, position sizing or transaction cost modelling.

Download the skill