Figgie Auto
An algorithmic sandbox for Jane Street's game, "Figgie"
Install / Use
npx skills add 0xDub/figgie-autoInstalls into whichever agent you are using.
README
Figgie Auto
An algorithmic sandbox for Jane Street's game, "Figgie". It's an excellent game that simulates "exciting elements of markets and trading". Coming from the HFT world, my strategy was to play the spread but due to laziness I figured I'd write up a program instead of learning the keybindings.
There is a lot to unpack in this game, especially when thinking about different types of market participants and how they operate. The most common route I'd imagine is to try and find the goal suit, which is the only suit that is awarded points each round. However, you could also go the route of providing liquidity to the participants, aiding in discovery. Whatever the case, I hope that the journey is fun
<hr>Development
event_driven: This type of player makes a decision on each update. Possible branch of strategies fall under HFTgeneric: This player makes a decision once every few seconds (adjustable inmain.rs). It's akin to a QR's setup
You can find barebones examples for both in the player folder.
Current Players
TiltInventory: On being dealt a hand, it finds the highest card and assumes it's the common suit. It semi-aggressively bids on the predicted goal suit market while selling all their other cardsSpread: A dumb market-maker, placing a wide quote range on all cards, attempting to profit off the order flowSeller: Quite conservative and defensive one, it attempts to sell all of its inventory and make up the cost of the anteNoisy: Consider this one as retailPickOff: An event-driven, opportunistic player - picking up cheap inventory in an attempt to sell it at a later priceTheHoarder: The goal for this strategy is to amass 6x of each card to mathematically guarantee a win and secure the pot. High risk, low reward, yet the pitfalls are quite insightfulPrayingMantis: A byproduct ofTheHoarder's pitfalls; likeSellerit attempts to offload it's inventory then aggressively buys up inventory of the perceived goal suit, based on last trade price. It has its own insightful pitfalls as well
To Jane Street: On the off chance you read this, I'd like to say I appreciate this game. If you'd like me to take it down, I completely understand and feel free to reach out
Inspired by: https://www.figgie.com/index.html
Related Skills
clawhub
385.5kSearch ClawHub for skills when a requested capability is not already available; install, verify, update, uninstall, publish, or sync skills.
coding-agent
385.5kDelegate coding work to Codex, Claude Code, or OpenCode as background workers; not simple edits or read-only code lookup.
node-connect
385.5kDiagnose OpenClaw Android, iOS, or macOS node pairing, QR/setup code, route, auth, and connection failures.
taskflow
385.5kCoordinate multi-step detached tasks as one durable TaskFlow job with owner context, state, waits, and child tasks.
