Skip to content

Commit 33504f2

Browse files
committed
Update HuskyBench (Poker) prompts
1 parent ba1738d commit 33504f2

3 files changed

Lines changed: 37 additions & 6 deletions

File tree

configs/pvp/huskybench.yaml

Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
tournament:
2+
rounds: 25
3+
game:
4+
name: HuskyBench
5+
sims_per_round: 1000
6+
players:
7+
- agent: mini
8+
name: p1
9+
config:
10+
agent: !include mini/default.yaml
11+
model:
12+
model_name: openai/gpt-5-mini
13+
- agent: mini
14+
name: p2
15+
config:
16+
agent: !include mini/default.yaml
17+
model:
18+
model_name: "anthropic/claude-sonnet-4-20250514"
19+
model_kwargs:
20+
temperature: 0.0
21+
prompts:
22+
game_description: |
23+
You are a software developer ({{player_id}}) competing in a coding game for Poker.
24+
In this game, you will write code to control a poker-playing bot, aiming to outsmart your opponents and win chips.
25+
Victory comes from crafting clever strategies—bluffing, reading opponents, and managing your chip stack effectively.
26+
27+
The game is played in {{rounds}} rounds. For every round, you (and your competitor) edit program code that controls your bot. This is round {{round}}.
28+
After you and your competitor finish editing your codebases, the game is run automatically.
29+
30+
Your task: improve the bot in `warriors/warrior.red`, located in {{working_dir}}.
31+
{{working_dir}} is your codebase, which contains both your bot and supporting assets.

configs/test/players/2/huskybench.yaml

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -10,9 +10,9 @@ players:
1010
name: p2
1111
prompts:
1212
game_description: |
13-
You are a software developer ({{player_id}}) competing in a coding game called CoreWar.
14-
CoreWar is a programming battle where you write "warriors" in an assembly-like language called Redcode to compete within a virtual machine (MARS), aiming to eliminate your rivals by making their code self-terminate.
15-
Victory comes from crafting clever tactics—replicators, scanners, bombers—that exploit memory layout and instruction timing to control the core.
13+
You are a software developer ({{player_id}}) competing in a coding game for Poker.
14+
In this game, you will write code to control a poker-playing bot, aiming to outsmart your opponents and win chips.
15+
Victory comes from crafting clever strategies—bluffing, reading opponents, and managing your chip stack effectively.
1616
1717
The game is played in {{rounds}} rounds. For every round, you (and your competitor) edit program code that controls your bot. This is round {{round}}.
1818
After you and your competitor finish editing your codebases, the game is run automatically.

configs/test/players/4/huskybench.yaml

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -14,9 +14,9 @@ players:
1414
name: p4
1515
prompts:
1616
game_description: |
17-
You are a software developer ({{player_id}}) competing in a coding game called CoreWar.
18-
CoreWar is a programming battle where you write "warriors" in an assembly-like language called Redcode to compete within a virtual machine (MARS), aiming to eliminate your rivals by making their code self-terminate.
19-
Victory comes from crafting clever tactics—replicators, scanners, bombers—that exploit memory layout and instruction timing to control the core.
17+
You are a software developer ({{player_id}}) competing in a coding game for Poker.
18+
In this game, you will write code to control a poker-playing bot, aiming to outsmart your opponents and win chips.
19+
Victory comes from crafting clever strategies—bluffing, reading opponents, and managing your chip stack effectively.
2020
2121
The game is played in {{rounds}} rounds. For every round, you (and your competitor) edit program code that controls your bot. This is round {{round}}.
2222
After you and your competitor finish editing your codebases, the game is run automatically.

0 commit comments

Comments
 (0)